mim Dataques GAD/CAM/CAE Electronic Design Automation Applications

mim Dataques GAD/CAM/CAE Electronic Design Automation Applications

DataQuest

GAD/CAM/CAE

Electronic Design Automation Applications

mim

Dataques

Dataquest Incorporated

1290 Ridder Park Drive

San Jose, CA 95131-2398

(408) 437-8000

Telex: 171973

Fax: (408) 437-0292

United Kingdom

Dataquest UK Limited

Roussel House,

Broadwater Park

Denham, Nr Uxbridge,

Middx UB9 5HP

England

0895-835050

Telex: 266195

Fax: 0895 835260-1-2

Japan

Dataquest Japan limited

Shinkawa Sanko Building

2 Fl

1-3-17 Shinkawa

Chuo-kuTokyo 104

Japan

011-81-3-5566-0411

Telex: 781-32768

Fax: 011-81-3-5566-0425

France

Dataquest Europe SA

Tour Galli^ni 2

36, avenue du G6n6ral-de-Gaulle

93175 Bagnolet Cedex

France

(1>48 97 31 00

Telex: 233 263

Fax: (01)48 97 34 00

Korea

Dataquest Korea

Dacheung Building Room 1105

648-23 Yorksam-dong

Kangnam-gu, Seoul 135-80

Korea

011-82-2-552-2332

Fax: 011-82-2-552-2661

Germany

Dataquest GmbH

Kionstadter Strasse 9

8000 Munich 80

West Germany

Oil 49 89 93 09 09 0

Fax: Oil 49 89 930 3277

Dataquest Incorporated

Ledgeway/Dataquest

The Corporate Center

550 Cochituate Road

Framingham, MA 01701

(508) 370-5555

Fax: (508) 370-6262

The content of this report represents our interpretation and analysis of information generally available to the public or released by responsible individuals in the subject companies, but is not guaranteed as to accuracy or completeness. It does not contain material provided to us in confidence by our clients.

This Information is not furnished in cotmection with a sale or offer to sell securities, or in connection with the solicitation of an offer to buy securities. This firm and its parent and/or their officers, stockholders, or members of the fomilles may, from time to time, have long or short position in the securities mentioned and may sell or buy such securities.

Printed in the United States of America. All rights reserved. No part of this publication may be reproduced, stored In retrieval systems, or transmitted, in any form or by any means—mechanical, electrons, photocopyitig, duplicating, microfilming, videotape, or otherwise—without the prior permission of the publisher.

®1991 Dataquest ItKxsrporated

Welcome to Dataquest

CAD/CAM/CAE

You are in the

Electronic Design Automation Applications

binder

This binder contains detailed market statistics, including market share and forecasts, of the overall electronic design automation applications market.

It also contains Dataquest Perspectives dedicated to the electronic design automation applications industry segment.

Other* CAD/CAM/CAE binders are:

Dataquest Perspective

A series of multitopic publications that provide analysis on worldwide G\D/CAM/CAE markets, companies, technologies, exhibitions, and trade shows, as well as news and views from around the mdustry.

Source: Dataquest

A biannualiy updated collection of reference and statistical documents about the worldwide

CAD/CAM/CAE industry. Worldwide, regional, and industiy segment market statistics. Company

Backgrounders, and several guides such as How to Use Dataquest, Dataquest Research Methodology,

and Dataquest High-Technology Cuide—Segmentation and Glossary are contained in this binder.

Mechanical Applications

Contains detailed market statistics, including market share and forecasts, of the overall mechanical applications market It also contains Dataquest Perspectives dedicated to the mechanical applications industry segment.

AEG and C I S Applications

Contains detailed market statistics, including market share and forecasts, of both the architecture, engineering, and construction CAD and geographical information systems market. It also contains

Dataquest Perspectives dedicated to the AEC and CIS applications industry segment

C A D / C A M / C A E Europe

Contains detailed market statistics, including market share and forecasts, of both the overall European market. It also contains Dataquest Perspectives dedicated to the European CAD/CAM/CAE industry.

C A D / C A M / C A E Asia

Contains detailed market statistics, including market share and forecasts, of the overall Asian market

It also contains Dataquest Perspectives dedicated to the Asian CAD/CAM/CAE industry.

Personal C A D and Distribution Channels

Contains detailed market statistics, including market share and forecasts, of the overall personal CAD market. It also contains Dataquest Perspectives dedicated to the personal CAD industry segment as well as a Dealer Directory and analysis of the indirect distribution channel.

m

*AII CAD/CAM/CAE subscribers receive the Dataquest Perspective and Source: Dataquest binders, inquiry privileges, and access to Dataquesfs Information Resource Center. In addition, subscribers to individual product or geographical segments receive the segment-specific binders: Mechanical Applications, Electronic Design Applications, AEC and

CIS Applications, Personal CAD and Distribution Channels, Europe, and Asia. Each regional service and segment also provides Dataquest Perspective and Source: Dataquest coverage unique to that region or segment For more information, contact your Dataquest marketing representative at (408) 437-8000.

CAD/CAM/CAE

Electronic Design Automation Applications

Table of Contents

Market Statistics

Market Share

Forecast

Dataquest Perspectives

Quarterly index

Dataquest Perspectives covering the electronic design automation industry segment

Dataquest

V W n aconnanyof

A l i i ThcDun&BradsticctCorpofation

Dataquest

Perspective

I N F O R M A T I O N R E S O U R C E C E N T E R

D A T A Q U E S T I N C O R P O R A T E D

1290 Rldder Park Drive

San J o s e , C A 95131-2398 t408) 437-8600

CAD/CAM/CAE

Electronic Design Automation

Vol. 1, No. 3 December 30, 1991

Market Analysis

IC Layout Market Continues to Stumble

In 1991, the IC layout market for electronic design automation (EDA) software experienced little, if any, growth, thereby continuing a pattern that has evolved in the past several years. In spite of this situation, however, Dataquest believes that some growth opportunities remain in the IC layout arena. This research provides IC layout market share data and examines in detail the forces, trends, and issues affecting the market currently and into the next five years.

By Ron Collett Page 2

CAE Market Anatomy Reveals Groivtb Opportunities

Computer-aided engineering (CAE) products are maturing as most design products introduced over the span of a decade reach the end of their life cycle. Dataquest believes that the time is ripe for EDA vendors to seize an opportunity to offer new CAE tools to a market anxiously av\^aiting their arrival. This research offers a dissection of the CAE market, including a comparison of the market's various segments that are reaching technological obsolescence and market saturation.

By Ron Collett and Robert Beachler Page 9

lochnology Analvsis

MCM Design Tools Take Over Layout Growth Leadership

Multichip modules (MCMs) are emerging as an important packaging technology, which should succeed surface-mount technology as the next important packaging technology in

Europe. This Dataquest research defines MCMs and the customers who will buy MCM tools from EDA vendors while assessing the overall outlook for MCM. design tools market in

Europe.

By Jim Tully Page 14

Nfvvs tine! Views

Mentor Graphics and Siemens Nixdorf Enter Marketing Partnership

By Jim Tully Page 19

©1991 Dataquest Incorporated / 1290 Rldder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAD/CAM/CAE—Electronic Design Automation

Market Analysis

IC Layout Market

Continues to Stumble

N u m m a r y

Dataquest's research indicates that the IC layout software market grew to $182.7 million in 1990, an 11.7 percent increase over 1990's $164 million figure. Preliminary analyses suggest that the market experienced little, if any, growth in

1991. Despite the market's sluggishness during the past several years, Dataquest believes that several growth opportunities remain. These include floor-planning tools, automatic placement and routing systems, and the tools and interfaces necessary to support timing-driven layout.

Our analyses further conclude that the IC layout software industry is in its final stage of evolution, having weathered a period of relentless consolidation during the past five years. The surviving set of players, most notably Cadence

Design Systems Inc. and Mentor Graphics Corporation, will now define the directions that the

IC layout business follows in the next several years. However, the overall business opportunity will not be governed solely by the collective market power wielded by these vendors. The market, as always, will have the final word. In fact, we expect the high level of concentration in the IC layout industry to give rise to a new generation of small niche suppliers, which in turn will provide customers with an alternative to the current titans. These niche IC layout players will be given a chance to enter the market, provided they can develop a technically

superior solution that can be easily alxorbed

into the customer's design environment.

Table 1

IC Layout Market Forecast

(Millions of Dollars)

The IC layout market is currentiy soft, a byproduct of the overall malaise in the electronics industry. However, in our view, once the electronics industry recovers from its current downturn, the IC layout industry will once again move into a period of growth. We believe that growth will likely result from providing physical design tools and technologies to a somewhat new kind of customer, the electronic systems design market.

However, traditional IC layout markets, such as the captive, merchant, and fabless semiconductor suppliers, all of which depend heavily on merchant tools, will continue to require advanced chip layout technologies. In addition, as more systems manufacturers begin taigeting the high-volume consumer electronics market, they will quickly realize the additional profitability that can be achieved by reducing die size.

Application-specific standard IC makers will also be impacted by systems manufacturers' attempts to target the consumer marketplace. These companies will attempt to design more chips in a shorter amount of time and reduce die size as much as possible. Table 1 shows that the market is expected to grow at a compound annual growth rate (CAGR) of 15 percent between

1990 and 1995. This research provides the following information:

• IC layout market forecast

• 1990 market share data

• IC layout market forces, trends, and issues

• IC layout submarket analysis, forecast, trends, and issues

IC Layout Market: Yesterday and Today

During the 1980s, the IC layout industry has gone through four stages of evolution: oligopoly, fragmentation, consolidation and shakeout,

Software

Hardware

Service

Tota]

1990

182

191

81

Wi

Source: Dataquest (December 1991)

1991

186

204

84

m

1992

208

214

92

514

1993

263

258

104

888

1994

312

284

118

1,026

1995

364

315

132

811

CAGR (%)

1990-1995

15

10

10

12

©1991 Dataquest Incoiporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAD/CAM/CAE—Electronic Design Automation and, again, oligopoly. From the late 1970s to early 1980s, the industry could have been characterized as an oligop>oly, dominated by just a few large players, namely, Applicon and Calma. By the mid-1980s, these companies were losing momentum and were unable to adapt to the market's demand for new technology. A myriad of suppliers began entering the market to fill the growing void.

Indeed, by 1986, no less than 20 vendors were competing in the hotly contested IC layout software business. Examples include Caeco Inc.,

Calma, Control Data Corporation, Daisy Systems

(now Dazix), ECAD Inc., Mentor Graphics,

Racal-Redac Group Limited, Scientific (Calculations, SDA, Seattle Silicon Corporation, Seiko

Instmments U.S.A. Inc., Silicon Compiler Systems Corporation, Silicon Design Labs, Silvar-

Lisco, Tangent, Valid Logic Systems Inc., and

VLSI Technology Inc. The result was a highly fragmented industry, which the market was both unable and unwilling to support. What followed was the industry's inevitable shakeout and consolidation period, which has been under way since 1985. The most recent evidence of consolidation is (Cadence's merger with Valid Logic, which was consummated at the end of December 1991.

Today's IC layout industry is once again an oligopoly, with Cadence Design Systems and Mentor Graphics together capturing 73 percent of the 1990 IC layout software market. Upon consummation of the Cadence-Valid merger, the combined entity together with Mentor Graphics captured 78.6 percent market share in 1990.

Table 1 shows that the top four IC layout software suppliers—which includes Cadence,

Mentor, Seiko, and Valid Logic—^held 87.5 percent of the IC layout software market in 1990.

Table 2 illustrates Cadence's firm grip on the

IC layout software market. The company's

55.7 percent market share in 1990 represents an increase of 11.5 percentage points over

1989. Cadence's growth has come at the expense of several competitors, most notably

Mentor Graphics, which experienced a 5.5-point drop in 1990. Valid Logic and Seiko Instruments also lost market share in 1990.

Forces Acting o n the

IC Layout Market

Overview

Despite the success that Cadence Design Systems has enjoyed during the past few years, the IC layout market has experienced relatively mild growth. Reasons can be traced to several sources. For instance, in the latter part of the

1980s, the IC layout industry endured enormous volatility, as witnessed by the myriad of mergers and acquisitions. This caused the market to view the viability of many fledgling commercial

IC layout sofi^vare suppliers with uncertainty and doubt. Moreover, as a by-product of acquiring and merging with other companies, many

IC layout suppliers were sidetracked by the challenges of absorbing other companies (such as integration of product lines, distribution channels, facilities, and corporate cultures).

More recently, the IC layout industry has fallen victim to the overall slowdown in the semiconduaor industry, which traditionally has represented the lion's share of the IC layout market. Indeed, in 1990, approximately 50 percent of the revenue captured by IC layout suppliers was generated by sales into the merchant semiconductor industry. The semiconductor

Table 2

1990 IC Layout Market Share

Cadence

Mentor

Seiko

Valid

Silvar-Lisco

Others

Total

Total

120.9

75.9

35.6

19.0

10.8

193.5

455.7

Revenue ($M)

Hardware

0

22.0

15.3

5.3

0

148.8

191.4

Software

101.7

31.9

16.2

9.9

7.0

16.0

182.7

Total

26.5

16.7

7.8

4.2

2.4

42.4

100

Market Share (%)

Hardware Software

0

55.7

11.5

8.0

2.8

0

77.7

100

17.5

8.9

5.4

3.8

8.7

100

Note: The Others category includes hardware-only vendors such as Sun, DEC, HP, and IBM; the Total category includes hardware, software, and maintenance revenue.

Source: Dataquest (December 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAD/CAM/CAE—Electronic Design Automation industry's slowdown has been due laigely to the downturn in the personal computer market, as well as other related industries. Figure 1 shows the 1990 IC layout market segmented by industry.

Capital Budgets for EDA in the

Semiconductor Industry

Dataquest's research indicates that capital budgets for electronic design automation (EDA) tools within the semiconductor industry as •well as the systems design market have expanded only marginally in the past several years. For instance, in 1987, the average EDA budget of a

U.S. semiconductor supplier was approximately

$589,000; in 1988, this figure grew to $788,000; in 1989, the average budget swelled to approximately $1.1 million.

Since then, however, U.S. EDA expenditure has been leveling off or even declining. For instance, Dataquest's most recent data indicate that the average EDA budget among semiconductor manufacturers reached approximately

$906,000 in 1991. Dataquest projects that capital budgets for EDA tools within the worldwide semiconductor industry will grow an average of 8 percent in 1992. Most of this growth will be in the Asian market.

Despite the fact that average capital EDA budgets among semiconductor suppliers are expanding modestly, the overall market for IC layout software is extending beyond traditional kinds of semiconductor manufacturers. New market segments include systems manufacturers that previously have not ventured into the world of

IC layout, as well as fabless semiconductor design companies, whose number is steadily growing. We anticipate that many systems manufacturers will initially extend their design methodologies to include floor planning.

Whereas others will expand into both floor planning and full-blown IC layout. Nearly all systems manufacturers will adopt timing-driven layout technologies once they become available, provided the products are easy to use and are useful in a production design environment.

Increasing Industry Stability

Between 1986 and 1987, the IC layout software market grew 58 percent, from $105 million to

$l66 million. This growth was in response to the void left by the giants of the time, namely

Calma and Applicon. It was also a reflection of the rapidly growing ASIC design market, which gave birth to ASIC manufacturers, a new class of IC layout software customer.

However, from 1987 through 1991, the IC layout industry experienced enormous volatility, with an array of mergers and acquisitions precipitated by the industry's fragmentation.

Dataquest believes that this enormous level of merger and acquisition activity contributed heavily to the industry's sluggish growth of less dian 4 percent CAGR between 1987 and 1991.

During this period, the market expanded from

Figure 1

Worldwide IC Layout Software End-User Market Segmentatloii—1990

Others 5.2%

Industrial Control 1.8%

Automotive 3.0%

Milltaiy/Aerospace 4.4%-

Telecommunications 7.1%

SOUKS: Dataquest (December 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAD/CAM/CAE—Electronic Design Automation

$160 million to an estimated $186 million in 1991.

Consolidation left a trail of uncertainty in its wake. Moreover, as a by-product of acquiring and merging with other companies, many suppliers were sidetracked by the integration challenges of absorbing other companies and, thus, were forced to divert resources from product development, marketing, and sales.

Just as fragmentation and instability created IC layout market uncertainty, Dataquest anticipates that the IC layout industry's increased concentration and stability will help fuel market growth during the next three years. This conclusion stems from our belief that the new oligopoly will engender market confidence in suppliers, creating an attractive climate for chip developers to purchase more merchant

IC layout took. However, Dataquest believes that a range of large chip manufacturers that have captive IC layout software development capabilities will not acquire merchant products until those products demonstrate a marked advantage over internally developed tools.

Industry stability will be further strengthened by the various technology partnerships and increasingly tighter supplier/vendor relationships that have formed among merchant IC CAD companies, semiconductor suppliers, and systems manufacturers. A recent example is the parmership foiged between Cadence and Fujitsu

Microelectronics Inc.

Ramifications of Concentrated

Market Power

It has become dear that IC layout industry control will rest primarily in the hands of Cadence and Mentor, at least for the next two to three years. Dataquest believes that such enormous market power wielded by only two players will cause the market to enter a period of polarization. Under the new industry structure, there will be a limited supply of substitute products, the bargaining power of the customer will diminish, competition among industry rivals will be less intense, and the threat of new entrants will be minimal. Thus, customers in both the semiconductor and systems industries will be forced to line up with either Cadence or Mentor. Many, in fact, are likely to share their internal development strategies and technologies with one or the other, jointly defining and/or developing the next generation of IC layout technologies.

In essence, Mentor and Cadence will become extensions, albeit attenuated, of the customer's back-end IC layout team. Some customers will certainly hedge their allegiance to either one or the other by straddling the line between the two, but these customers will end up simply following the trends defined by the tighter alliances.

Demand in the Face off a Semiconductor Industry Dotvntum

Dataquest's research indicates that the IC layout market is becoming more selective with respect to large capital expenditure on IC layout tools.

In short, decisions to invest several hundred thousand dollars in a single IC layout "seat" are being weighed more carefully than ever before.

Dataquest's research further suggests that the

North American IC layout market is currendy in a period of saturation, with most users indicating that they do not plan to increase the number of IC layout seats.

However, Dataquest believes that ASIC designers will be willing to pay a premium for tighter integration between the front end of design and layout. This tighter integration will come in different forms. For instance, most of the ASIC design market will gravitate toward using floor planning and timing-driven layout tools.

In Asia, where semiconductor development and manufacturing continues to expand, a significant

IC layout opportunity remains. This opportunity runs the gamut from replacing outdated merchant IC layout stations to replacing internally developed tools. Also, in Asia just as in Nortii

America, we believe that there will be significant demand for next-generation IC layout tools that improve silicon efficiency and speed up the entire layout process.

Pricing Pressure

Although opportunity remains, the market will be putting increased pricing pressure on IC layout suppliers. In short, the recent commoditization of many electronics products is casting a sobering light on the prices that electronics manufacturers can charge for their products.

Examples that immediately come to mind include the notebook computer and cellular telephone markets, where prices are rapidly falling. Reduced prices are cutting deeply into profit margins, which in turn will continue to impact capital expenditure on EDA tools and, specifically, IC layout software.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAD/CAM/CAE—Electronic Design Automation

Tools and Technologies in Demand

Examples of tools and technologies that will be in high demand include floor planning, symbolic editing, compaction, layout synthesis, three- and four-layer metal automatic placement and routing tools, advanced polygon editors, robust design verification tools, and library development tools. Much of the existing internally developed IC layout technology is being rapidly surpassed by the capabilities of the merchant tools, and many chip suppliers, especially in Asia, have no choice but to turn to merchant IC layout products. This has already begun in earnest.

Rumors of "Free" Silicon Are

Greatly Exaggerated

Dataquest believes that the demand will increase for IC layout tools that improve silicon efficiency. (Certainly one could argue that with the price-per-gate of silicon on standard ICs, gate arrays, and cell-based ICs continuing to spiral downward, the market w^ill be less concerned with silicon efficiency. This, in turn, would appear to reduce market demand for IC layout tools that optimize silicon efficiency.

After all, the additional cost paid for less efficient utilization is ostensibly marginal. Dataquest believes that this argument is a seductive but dangerous trap for the unwary.

In our view, silicon efficiency will become increasingly important throughout the decade.

Our belief rests on the assumption that electronics manufacturers will attempt to develop more products for the consumer electronics market. Electronics manufacturers will turn toward the consumer market because they are recognizing that existing markets are rapidly saturating. Thus, many manufacturers believe that serving the consumer market is a viable and necessary strategy for growth. To serve this market, electronics manufacturers will be merging computing and telecommunications technology with consumer electronics.

The consumer market boasts significantly higher volume opportunities than do the traditional computer markets. In high-volume markets such as consumer electronics, shaving even a small amount of silicon area can add significantly to bottom line profitability.

The relative importance of silicon efficiency to electronics manufacturers dep)ends, of course, on product strategy. Product strategy is a function of the relationship between market window, market maturity, and product pricing.

For example, time to market is dearly becoming a more influential factor that will determine a product's chance for market success. Early entrants to a particular market segment can charge a premium because of the limited supply. Thus, minimizing product development time and launching a product at the leading edge of the market window is often most important.

Cost of goods, such as ASIC and applicationspecific standard IC costs, and thus silicon efficiency, is less important at the early phase of a bui;geoning market opportunity.

For example, consider the pen-based computer market. It is just beginning to unfold, enabling early entrants such as GRiD Systems Corporation, Momenta Coiporation, and NCR Corporation to take advantage of the minimal competition by charging higher prices and creating brand recognition. Price is less of an issue for early adopters of the product. In just a few years, however, as the technology moves into the mainstream market, price will become a significant differentiator among competitors. Consequendy, reducing silicon area within the system will be among the design goals for the nextgeneration products.

In sum, silicon inefficiency will only be tolerated when it can be offset by other sources of value. To verify this, ask a consumer electronics manufacturer whether silicon is "free."

Workstations, Frameworks, Open

Systems, and Standards

As Dataquest indicated last year, the skyrocketing integer and floating-point performance of the new generation of workstations and servers continues playing a positive role in the IC layout market's future growth. Workstations churning away at 30 to 50 million instructions per second (mips) and 6 to 8 million floating-point operations per second (mflops) are capable of handling most compute-intensive IC layout tasks.

However, chip manufacturers continue to complain about long run times for simulation, layout, and design verification tasks. The reason, of course, stems from the fact that chip complexity continues to grow. This growth further taxes IC layout software and thus offsets many of the performance gains achieved with the new generation of computing technology.

Frameworks, open systems, and standards will also play a role in whetting the market's appetite for IC layout products. TTie ability to

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAD/CAM/CAE—Electronic Design Automation integrate point tools more easily into the emerging frameworks and open systems creates a powerful draw for many potential buyers. Moreover, the combined market power of Cadence/

Valid and Mentor Graphics, together with the efforts of the C;AD Framework Initiative, has the potential to speed the proliferation of standards.

On the other hand, the fact that market power has been funneled into the hands of only two players. Cadence and Mentor, may also act to slow the standards efforts. The direction that standards will take will be apparent by mid-1992. However, one thing is certain: The market will not expand much until standards enable customers to more easily interchange tools from different suppliers.

Dataquest's research indicates that the market is not expecting to pay much for integration capabilities, such as those emerging from merchant framework products. Indeed, the market expects this feature to be intrinsic to the application software, not something that has an additional price tag.

IC Layout Software Submarket

Analysis

Figure 2 segments the various submarkets that make up the IC layout software market.

FuU-Custom IC Design Tools

With total sales of $55.1 million, full-custom

IC design tools represented 30.3 percent of the market, second only to design verification, which was the largest segment in 1990. With

1989 sales of $54.2 million, market growth in

1990 was negligible. Products in this segment include polygon editors, symbolic editors, compaction tools, and floor-planning systems.

Although this market is highly penetrated by polygon editors, many of these installed polygon-editing systems are based on older technology such as Calma's GDS II. Dataquest expects nearly all of these seats to be replaced in the next two years by newer, more advanced polygon editors, as well as symbolic editors and compaction tools.

Although cell-based and gate array design continues to replace the handcrafted full-custom approach, we believe that nearly all chipmakers will still need polygon editors. These tools will be used to develop macrocell libraries as well as make adjustments to the layout performed by the automatic placement and routing systems. We anticipate that polygon editing will be done less and less in the next five years because of the increasing importance of reducing time to market. This will mandate the use of symbolic editing and compaction, which will

Figure 2

1990 Worldwide IC Layout Software Market Breakdown

Module Generation Development Tools 5.5%

Compilers 6.7%

Automatic Place and Route (Cell-Based IC) 11 -5%

Automatic Place and Route (Gate Aray) 12.2%

Source: Dataquest (December 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

8 CAD/CAM/CAE—Electronic Design Automation replace and/or complement polygon editing as the technology improves.

Dataquest believes that floor planning will also find widespread use during the next few years as systems manufacturers designing ASICs want to develop the chip's floor plan to ensure that timing performance is optimized. Systems designers developing ASICs represent a new market for IC layout vendors, as these engineers traditionally have not been directly involved in performing ASIC layout. This task has been the responsibility of the ASIC manufacturer.

We believe that the shift in floor-planning responsibility will be a significant trend in ASIC design in the next few years. In fact, by 1995, we expect the majority of ASIC designers to perform floor planning. Dataquest forecasts that the full-custom IC layout tool market will experience a CAGR of 10 percent from 1990 through 1995.

Automatic Placement and Routing

Dataquest expects the market for automatic placement and routing to experience very robust expansion during the next five years.

We expect the gate array and cell-based automatic placement and routing markets to have a CAGR of 31 and 19 percent, respectively.

Figure 2 shows that the combined segments represented 23.7 percent of the overall IC layout software market.

Fueling this sector's g i b ^ h is the fact that both

ASIC and standard IC manufacturers will be forced to adopt advanced layout technologies to meet the challenges of laying out a rising number of increasingly complex semicustom chips.

For instance, Dataquest forecasts that, by the end of 1994, more than 75 percent of new

CMOS gate array design starts will be more than 10,000 gates. Examples of new IC layout technologies include three- and four-layer metal placement and routing systems, layout synthesis systems, and advanced compaction tools.

Similar to the floor-planning trend, we expect an increasing number of leading-edge systems manufacturers to become more involved in the layout of their ASICs. In the past, very few

ASIC customers have performed layout; indeed, this has been the almost exclusive territory of the ASIC foundry. We believe that this trend will be driven by the following forces:

• Systems houses building high-speed systems wiU want to get actual, albeit preliminary, interconnect delay information as early as possible in the design cycle, as well as on a continuous basis throughout the design cycle.

Performing the layout in-house enables and facilitates this feedback of delay data to the

ASIC design team.

• As the profit margins on electronic hardware narrow, electronic systems manufacturers will want to perform the layout as a means of reducing die size and, therefore, combat profit margin erosion.

• As time-to-market pressures become more severe, systems houses will want to ensure that their products get to market on time, and taking control over the complete design cycle (assuming the systems house has brought in competent layout expertise) can help.

• Systems houses want to maximize chip performance, and performing the layout internally allows them to optimize timing performance.

Design Verification

With $61.5 million in sales, the design verification market was the largest segment of the IC layout market in 1990. The market grew approximately 36 percent between 1989 and

1990. Dataquest expects the design verification market to experience a CAGR of 14.3 percent from 1990 through 1995. Our projections are based on the following assumptions:

• Chip manufacturers will replace lastgeneration technology.

• More systems manufacturers will begin performing ASIC layout.

• Mentor Graphics will begin to attack the market with its Checkmate product more aggressively.

• New chip companies will emerge (for instance, those that will develop applicationspecific standard ICs).

Cadence's Dracula product line has enjoyed a near monopoly in the design verification market for the past five years. Mentor Graphics had hoped to change this situation with its Checkmate design verification system. Checkmate was acquired via Mentor's 1S>90 acquisition of Silicon

Compiler Systems. Mentor Graphics has stated publicly that it plans to employ an aggressive pricing strategy to attack Cadence's stronghold.

Despite the prospect of dramatically lower pricing, displacing Dracula has become a formidable challenge for Mentor Graphics. The strategy has yet to bear fruit.

©1991 Dataquest Incoipoiated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

i i i

CAD/CAM/CAE—Electronic Design Automation

Design verification is among the most critical steps in the chip design process, and injecting a change in verification tool technology is a risky proposition, as it could cause a severe perturbation in the design cycle. Nonetheless,

Dataquest believes that Mentor will capture a piece of the verification market in 1992.

Module Generation Development

Tools and Compilers

With $22.4 million in sales, this market occupied approximately 12.3 percent of the overall IC layout market in 1990. The market was down considerably from 1989, when it recorded combined sales of $36 million. We had expected module-generation tools to enjoy continued market success, but insufficient marketing and product positioning by vendors caused the market to collapse.

Our research indicates that customers do not perceive that the current generation of products offer enough additional benefit to justify the cost and time required to fully utilize the technology. In addition, various segments of the market perceive hardware-description-languagebased logic synthesis to be somewhat of a substitute vehicle for creating soft megafunctions.

In our view, proper positioning and aggressive marketing can reenergize the module generation development tool market.

The cell compiler market (for instance, those products from ASIC suppliers), which was

$15 million in 1989, fell to $12.3 million in

1990. As we expected, this market is not likely to experience any significant growth in the next five years. Most ASIC manufacturers that offer compilers chaige little, if anything at all, for their compilers. Instead, most bundle them with the library, which is often heavily discounted or given away at no charge. In addition, we believe that third parties selling compilers will continue to meet with only limited, if any, success because of stiff competition from ASIC suppliers.

Dataquest Perspective

Dataquest believes that despite the IC layout market's relatively flat growth in the past years, several opportunities remain in this seaor, especially in Asian countries such as

Japan and Taiwan. On the other hand, the

North American and European markets will likely remain soft throughout 1992 but are projected to improve in 1993 and 1994.

By Ron Collett

CAE Market Anatomy

Reveals Grotvth

Opportunities

Siumnary

Many of the computer-aided engineering (CAE) products introduced in the past 10 years are reaching the end of their technology life cycle.

Examples include schematic entry, gate-level simulation, and various test products. On the other hand, newly introduced products targeting leading-edge users are finding widespread market success. Logic synthesis, mixed-level simulation, and hardware, description language (HDL)based design entry are just a few that fall into this category.

Dataquest believes that the market is anxiously awaiting the arrival of new CAE technology that leapfrogs the current generation. One indication that the market is in a holding pattern is the fact that the CAE software market grew only

10.8 percent in 1990. Nearly all of this growth was driven by HDL-based logic synthesis.

Preliminary analysis of 1991 reveals similar figures. Table 1 segments all of the submarkets that make up the worldwide CAE software market. This research dissects the CAE market and compares segments slated for growth with those reaching technological obsolescence and market saturation.

Digital CAE Software Market

Remains Soft

Digital CAE software, which includes design entry, verification, logic synthesis, and test, grew a modest 6.9 percent in 1990. Dataquest's preliminary analyses suggest that the market grew approximately the same amount in 1991-

Probing into the individual submarkets reveals that this sluggishness is localized to specific areas. The following analysis examines the individual submarkets within the digital CAE software arena.

Next-Generation Digital

Design Entry

Dataquest separates the digital design entry market into thre% categories; schematic entry, high-level design entry, and libraries. Table 1 shows that the schematic entry market grew a negligible amount between 1989 and 1990.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

f

1 f

p

^0

V/i

O"

H* llj

00

Ig t^ i 8

(-»*^

5 g! rt

•fl

3

0

1

^x is

Table 1

Worldwide CAE Software R e v e n u e b y A p p l i c a t i o n a n d Market (Millions o f D o l l a r s )

Electronic CAE (Including Analog)

Distal Design

Design Entry

Schematic Entry

High-Level Design

Libraries

Design Verification

Simulation

Static TA

Logic Synthesis

Test Automation

ATVG

Design for Test/Test Uff^ Synthesis

Fault Simulation

Other Digital

Analog Design

Design Entry

Schematic

Libraries

Design Verification

Circuit Simulation

Mixed-Signal Simulation

Other Anaioz Desien

NM = Nim m^anmjjfiil

Source: Dataquest (Deconber 1991)

Total

1989

6.23

2.67

29.50

19.85

110.51

41.04

26.21

14.83

61.42

59.54

1.88

8.05

590.95

479.69

217.02

164.00

17.35

35.67

187.72

160.50

16.22

16.69

38.41

ASIC

283.34

258.95

83.10

65.62

4.82

12.66

117.21

104.18

13.03

13.69

26.89

4.61

2.10

20.18

8.16

24.39

9.71

4.28

5.43

14.20

13.51

0.69

0,48

Board

257.91

171.79

102.29

75.18

6.63

20.48

46.12

42.94

3.18

0

10.71

1.22

0.18

9.31

11.56

86.12

31.33

21.93

9.40

47.22

46.03

1.19

7.57

FPGA/PLD

0.01

2.99

0.81

0.40

0.40

0.01

0.87

NM

NM

NM

NM

NM

NM

NM

NM

49.70

49.70

31.64

23.21

5.89

2.54

13.39

13.38

Total

1990

654.65

512.66

244.24

165.83

36.64

41.77

178.31

163.80

14.51

41.02

30.72

6.95

2.45

21.32

18.38

140.90

57.36

39.02

18.34

73.55

64.04

9.51

9.99

ASIC

298.31

245.50

83.84

53.24

17.22

13.38

103.48

92.38

11.10

30.96

21.92

5.21

2.25

14.46

5.30

52.81

22.36

11.91

10.45

30.20

24.28

5.92

0.25

,11^

5ii

6

CAD/CAM/CAE—Electronic Design Automation 11

Moreover, Dataquest's end-user research indicates that this market is virtually saturated. This fact should come as no surprise to companies supplying products to this sector.

Schematic entry is last-generation technology that is used primarily for gate-level design.

Nearly all of the schematic entry systems on the market are geared toward bottom-up, not top-down, design. Conversely, the impending generation of schematic entry technology under development will focus on supporting the topdown design methodology. Several companies are likely to begin shipping such systems in

1992.

It is also not surprising to see that highlevel design entry, which primarily consists of

HDL-based entry, surged forward in 1990.

Table 1 shows that the market expanded from

$17.3 million in 1989 to $36.6 million in 1990.

"We anticipate that the market grew at least another 50 percent in the past year. During the next 12 months, we expect the market to grow^ at least another 50 percent. Indeed, our fiveyear projections (1990 to 1995) indicate that this segment will experience a compound annual growth rate (CAGR) of approximately

52 percent.

Fueling this market is the shift toward topdown design. Using an HDL instead of schematic entry for ASIC design boosts productivity several times. Despite the trend toward using an HDL in lieu of schematic entry, Dataquest believes that exclusive use of HDL for

ASIC design is a short-term aberration. We believe that designers will use a combination of HDL and graphical entry in the future. We believe that a graphical interface is far more intuitive than a language-based interface. The marketplace clearly prefers graphical entry, as evidenced by the fact that it continues to rank higher in importance than HDL-based entry.

On a scale of 1 to 5, with 5 being most important, users assign 4.1 (mean score) to schematic entry and a 3-7 (mean score) to HDL entry.

The next generation of graphical design entry technology will be a significant departure from the current generation of gate-level entry products. In our view, designers will work with VLSI blocks buik from HDL. Designers will reuse and modify existing HDL blocks as well as create new ones to meet the demands of the project at hand. The HDL functions themselves will be represented as graphical blocks, but designers will be given the freedom to quickly access and modify the blocks' internal

HDL. The HDL among the various blocks will be linkable, synthesizable, and thus capable of generating a gate-level implementation of the

ASIC design.

Libraries continue to be the most significant problem facing design teams. Indeed, companies in the market believe that libraries are the most important element of their CAE arsenals, yet the model market continues to stumble along.

Table 1 reveals that the market grew from

$35.7 million in 1989 to $41.8 million in 1990.

We believe that the sluggishness can be traced to the following sources:

• Users do not expect to pay much for models.

• The breadth of models required to meet the market's needs is so wide that it is difficult for third-party modeling companies to operate.

We do not believe that there is an appropriate business model to support a full-line supplier of

VLSI models. Several companies have tried, but none have met expectations. It is simply too difficult for one company to develop a full line of models that are bug-free. Moreover, support costs can be astronomical when compared with achievable gross revenue. In our view, thirdparty modeling companies can be successful only if they remain narrowly focused. Under that scenario, companies take on more of a

"consulting business" look.

We believe that semiconductor manufacturers that make a particular component are in the best position to develop and support models of their parts. However, this is not likely to happen on a broad scale in the near fiiture.

Most semiconductor manufacturers view model development and support for the merchant market as a losing proposition to its bottom line finandals.

Lack of models causes a severe problem for electronic design automation (EDA) vendors. It will potentially impede grov*^ in the full-system simulation market. Indeed, Dataquest's research indicates that 32 percent of the North American market is performing system-level simulation.

Dataquest projects that this figure will grow to approximately 56 percent in the next 12 to

24 months. Without adequate models, however,

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

12 CAD/CAM/CAE—Electronic Design Automation such growth is difficult to achieve. If the market finds that the existing suite of models is insufficient, we expect the full-system simulation market to expand only 5 to 10 percent during this time.

Design Verification

Dataquest segments the digital design verification software market into two categories— simulation and static timing verification. Simulation includes all categories of logic simulation, from switch-level only simulators to behavioral and mixed-level products. Both the simulation and static timing verification software markets experienced difficulties in 1990. The reasons can be traced to the vendors supplying these products as well as to the products themselves.

For example, in the simulation arena, three of the four leading CAE vendors were unable to deliver what the market was demanding: fast, accurate, easy-to-use simulation technology that supported behavioral modeling. Mentor

Graphics Corporation was struggling to finish the development of its Quicksim n product. Its

Quicksim I product had run out of steam and was perceived to offer neither adequate performance nor modeling capabilities. Meanwhile,

Dazix was on the verge of liquidation and was unable to convince the market to purchase many of its products. Valid Logic Systems Inc.'s long-time simulation product, Validsim, was no longer competitive, and the company had only recently (in mid-1990) unveiled its nextgeneration simulator, Rapidsim. The only company that had any success in the simulation market was Cadence Design Systems Inc., with its then new^ly acquired Verilog-XL simulator from Gateway Design Automation.

Mentor, Valid, and Dazix each lost significant momentum in the simulation market in 1990.

This loss carried over into 1991 and is likely to continue until at least early to mid-1992. The market perceives Cadence's Verilog-XL simulator as the technology leader.

However, leadership in the CAE business can be fleeting. In 1992, we expect most CAE suppliers to set their sights on both the Verilog-XL installed base and the rapidly expanding VHDLbased simulation market. In addition, several other CAE companies are entering the fray with w^hat appear to be increasingly credible product offerings. Vendors in this camp include Viewlogic Systems Inc., Synopsys Inc.,

Racal-Redac Group Limited, and Vantage Analysis Systems Inc. Yet Cadence's merger with

Valid will make penetration of the Verilog-XL and Valid installed bases more difficult for all companies.

Sluggishness in the simulation market in 1990 is a reflection of the fact that Table 1 does not separately break out the HDL-based, mixed-level simulation market. In other words, all gate-level simulation produas are lumped under the simulation category, and products in that category have reached the end of their technology life cycle. The market for such products is declining. HDL-based mixed-level simulation products are in high demand, as evidenced by the groAvth of the HDL design entry market in

1990. Dataquest believes that they will remain in high demand for the foreseeable future, paralleling the growth rate of the HDL-based design entry market.

Driving this growth will be the increasing acceptance of HDL-based design in both ASIC and board design, as well as more extensive library support for the leading HDLs (Verilog

HDL and VHDL). In addition, the market for simulation products is expanding. An increasing number of new simulators are being purchased to perform board- and/or system-level simulation. Board/system-level simulation revenue grew 38 percent between 1989 and 1990.

Rapid growth in this segment is being flieled by the increased availability of simulation models and improved performance of simulators, which is allowing designers to simulate complete boards and systems. Of course, the most significant driving force behind growth is the need to increase produaivity and quality levels in electronic design.

Meanwhile, the static timing verification market fell slightly in 1990. Products available to date have not been effective at separating erroneous timing problems from true timing problems.

Thus, the user is burdened with culling the nonexistent, yet highlighted, violations. This has made the products unattractive to large segments of the market.

In addition. Cadence's Veritime timing analysis tool has so far been unable to piggyback onto the Verilog-XL's success. The primary reason is the lack of ASIC libraries supporting the product. Static timing verification for board-level design remains a small market, because until recently only a small percentage of the overall market was performing any kind of board-level verification.

©1991 Dataquest Incorporated / 1290 Bidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAP/CAM/CAE—Electronic Design Automation 13

We believe that the static timing verification market can rebound, provided that suppliers can correct the problems with the current set of products. Certainly the technology is needed as indicated by the fact that the market assigns it a score (mean) of 3-9 in terms of importance, with 5 being most important. Factors that will help to expand this market include the following:

• Increasing complexity and performance of

ASIC and board designs demands better timing analyses.

• The impending shift toward timing-driven design and layout will require extensive reliance on timing analysis tools.

Logic Synthesis: The BIG Winner

Logic synthesis is clearly the rising star in the

CAE software market. The market expanded

146 percent in 1990, reaching $41.02 million in

1990. This amount includes logic synthesis for

ASICs, field-programmable gate arrays (FPGAs), and programmable logic devices. We expect this market to continue to experience very robust expansion in the next two to four years.

We had expected the ASIC segment of the logic design market to reach $60 million by the end of 1991. Our forecast in April 1990 was based on the assumption that the leading EDA vendors would have been able to establish themselves in this segment by early 1991. They have been unable to do so. To date, only Synopsys has been effective in penetrating this market.

We expect the logic synthesis industry to fragment considerably over the next two years, with no fewer than 10 suppliers entering this market during that time. The competition will become fierce, especially in the VHDL-based logic synthesis segment. Dataquest projeas that the logic synthesis market will have a CAGR of

55 percent during the next four years.

Test Automation: An

Untapped Well

Dataquest divides the test automation software market into three categories: automatic test vector generation (ATVG), design for test (DFT) plus test logic synthesis, and fault simulation.

The overall market fell from $38.4 million in

1989 to $30.72 million in 1990. We are projecting that the market will again be in the

$30 million to $40 million range in 1991.

Table 1 shows that the fault simulation segment was the primary contributor to the market's shrinkage in 1990. The military/aerospace industry typically exhibits the most interest in fault simulation tools. Military and aerospace electronics must conform to stringent fault coverage and fault isolation specifications. Despite the need for more fault-simulation capacity, the slowdown in defense spending is trickling through the defense industries, which in turn is impacting expenditure on fault simulation tools. We expect this slowing to continue.

Other market segments such as the computer and telecommunications also show a fair amount of interest in fault simulation tools, but the sense of urgency regarding fault simulation is less than in the mil/aero sector. This sector's fault coverage and isolation needs are more goal-driven and less sjjecification-driven.

EDA manufacturers have been unsuccessful at penetrating the ATVG, DFT, and test logic synthesis markets. Dataquest believes that the following factors have contributed to limited market growth:

• The tools and technologies offered to the market are expensive and limited in scope.

• The tools and technologies offered to the market have thus far not been integrated well into the design flows and environments of electronics manufacturers.

• The primary market potentially requiring these products is the ASIC design market (as opposed to the board design market), and the average size of ASIC designs has only recendy crossed the 10,000-gate mark. Thus, the need for advanced test tools has been limited.

• Except within the mil/aero market, test traditionally has not been built into the design methodologies of electronics manufacturers.

Dataquest believes that the demand for ATVG,

DFT, and test logic synthesis tools will increase as ASIC complexity expands. We expect ASIC gate densities to grow dramatically during the next three to five years. Increasing gate density will be a by-product of the advances made in logic synthesis, top-down design, the proliferation of HDL models, and increasing reuse of existing designs.

Dataquest projects that the test automation market as a whole will experience a CAGR of approximately 15 percent between 1990 and

1993, expanding to a 35 percent CAGR between

1994 and 1996.

©1991 Dataquest Incorporated / 1290 Rldder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

14 CAD/CAM/CAE—Electronic Design Automation

Board-Level CAE Software

Surges Ahead

The market segmentation in Table 1 indicates that the board-level CAE software market grew almost 16 percent between 1989 and 1990, whereas its ASIC counterpart fell slightly. The primary reason for the ASIC market's slide is saturation in the schematic entry segment and lack of supply of next-generation simulation products. We expect the ASIC segment to begin expanding once suppliers complete development of the next generation of HDL-based verification and design entry systems.

Both the design entry and verification markets are the driving forces behind board-level CAE growth. It is clear that the CAE-for-ASIC design revolution of the 1980s is continuing to spill over into the board design market. We expect the trend toward increased usage of CAE software for board design to continue during the next three years. Following on its heels will be the next-generation opportunity: electronic systems design automation (ESDA).

In contrast to the ASIC CAE software market, the programmable logic device (PLD) and FPGA design tool market continues to show healthy growth. Dataquest expects this trend to continue as support of FPGAs and PLDs migrate to the workstation platforms. It is dear that the increasing level of acceptance of FPGAs by the design community is translating to a greater need for sophisticated design tools that accommodate this new generation of complex PLDs.

Dataquest expects the FPGA/PLD CAE software market to reach approximately $150 million by the end of 1995, translating to a CAGR from

1990 to 1995 of approximately 24 percent.

Analog CAE Opportunity

Emerges—In Asia

Table 1 shows that analog CAE software growth outpaced digital CAE software growth by a factor of four between 1989 and 1990.

The market expanded from $110.6 million in

1989 to $140.9 million in 1990. Fueling this growth is the introduction of more advanced analog design tools.

Much of the market's growth has been in

Asia, where the market expanded from

$25.7 million in 1990 to $46.8 million in

1991. This increase in consumption brings

Asia's analog CAE market in line with its analog component consumption, which is slightly less than that of North America.

Dataquest anticipates that analog CAE software growth will continue to outpace its digital counterpart but will remain significantly smaller than the latter. The most significant opportunity within the analog CAE software market is the mixed-signal verification market. Table 1 shows that the market grew from $1.88 million in 1989 to $9-51 million in 1990. Dataquest's preliminary analyses suggest that the market grew another

50 to 100 percent in 1991. (For more information on the mixed-signal verification market, refer to the article entitled "Interpreting the

Mixed Signals of the Mixed-Signal Market," published in the CAD/CAM/CAE EDA Dataquest

Perspective, Vol. 1 No. 1.)

Dataquest Perspective

Although overall growth of the CAE software market seems to be somewhat stalled, Dataquest believes that a dear set of opportunities remains. Examples indude HDL-based design entry and logic synthesis, system-level simulation, analog CAE, and the FPGA/PLD CAE markets. However, we believe that the worldwide economic slowdown will continue to have a negative impaa on the CAE software market.

Our preliminary analyses suggest that the CAE software market grew approximately 11 percent in 1991. We expect the market to be somewhat stronger in 1992, especially in Asia where there is a high degree of pent-up demand. •

By Ron Collett and Robert Beachler

T e c h n o l o g y Analysis

MCM Design Tools

Take Over Layout

Growth Leadership

Introduction

The multichip module (MCM) is emerging as the most important packaging technology since surface mount and will find increasing use in all electronics sectors. The technology demands new electronic design automation (EDA) tools, and vendors are quickly rising to this opportunity. But what are MCMs, which tools are needed, and who will buy them? Also, who will be the early adopters of the tools and devices, and how will the technology affect system design.'

This research analyzes these questions and assesses the outlook for MCM design, with spedal emphasis on the European market.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAD/CAM/CAE—Electronic Design Automation

15

System Interconnect Problems

The interconnection of ICs possessing 400 or more pins using conventional printed circuit boards (PCBs) leads to one conclusion: The large board area occupied by PCB tracks can negate the high levels of integration achieved within the IC. Either laige areas of the board cannot be used for components because of the density of copper track or large numbers of layers must be used. Both scenarios are usually unacceptable. Further, as system speed increases and tracks start to behave as transmission lines, interconnection becomes a nightmare.

Packaging and interconnection of ICs account for a significantly higher proportion of equipment cost than the total IC cost. This cost is therefore a major consideration in equipment design.

Problems Answered: Enter the

MCM

MCMs answer these problems and are the next generation of packaging technology, beyond surface mount, that meets the future needs of systems designers. In general, MCMs eliminate individual IC packages in favor of bare dice mounted on a common substrate, leading to improved packing density and system performance (due to reduced interconnection delays and track impedances).

Mainframe computer companies are not strangers to this technology, with Cray Research

Figure 1

Worldwide MCM Market Growth Assembled Modules

Inc., Fujitsu Ltd., IBM Corporation, and NEC

Corporation introducing machines based on

MCM technology during the 1980s. Personal computer companies are also keenly interested in MCMs. For example, a complete Macintosh

SE computer was implemented on a 2.5 x

3.5-inch silicon substrate as a joint project between Apple Computer Inc. and Dow

Chemical Corporation.

Many other applications exist, especially in the automotive, microwave, high-speed digital, and high-power areas. Dataquest estimates that the market for MCMs will grow at a very fast rate, achieving a 109 percent compound aiuiual growth rate (CAGR) in the next five years

(see Figure 1).

MCM Technologies

MCM substrates have been categorized over a spectrum of technologies by IPC conmiittee IPC-

MC-790. (Note: When comparing interconnect densities, the normal metric is the length of interconnect material in centimeters per square centimeter of substrate per layer.) These substrates range from high-density PCBs to full silicon implementations and arc detailed as follows:

• MCM-L (Laminated)—^Fabricated using typical

PCB processes and materials; targeted at lowend applications requiring interconnect densities of 50 to 150 crn/sq cm at up to 100 to

200 MHz; favored by the automotive industry

Billions of U.S. Dollars

CAGR (1991-1996) =109%

4.5-j

4-i

3.5-i

3-i

2.5-:

2^

1.5-j

^.^"^

0.5-:

1

Source: Dataquest (December 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

16

CAD/CAM/CAE—Electronic Design Automation

because of its relatively low cost; typically uses chij>-on-board (COB) assembly, wirebonded to the substrate, then encapsulated for protection

• MCM-C (Ceramic)—Ceramic substrates (both cofired and low-dielectric constant ceramics) offering economical interconnect densities between 100 and 250 cm/sq cm; traditional thick film hybrid process; favored by aerospace and military seaors because of harsh environmental capabilities; COB and flip-chip assembly commonly used; metal encapsulation offers physical protection and improves

EMC screening

• MCM-D (Deposited)—^Deposited wiring and dielectric on silicon, ceramic, or metal substrates; uses thin film processes for interconnect densities of 200 to 4(K) cm/sq cm and above (Silicon substrates offer good thermal properties because the devices are also silicon and exhibit the same thermal expansion characteristics. MCM-D substrates are still under development. State of the art in wafer size is 8 inches. These wafers can only support one or two typical modules and are therefore limiting economical production.);

TAB and flip-chip assembly

Most MCMs today are of the MCM-C type and are used for computer, aerospace, and military applications involving thick film hybrid technology on ceramic substrates. A major deciding factor in the choice of MCM technology for a given application is the cost-per-interconnect density. For low densities (less than approximately 50 cm/sq cm), MCM-L offers the lowest cost. For densities that are approximately between 100 and 200 cm/sq cm, MCM-C has the lowest cost. MOI-D is expected to offer lower cost at interconnect capacities that are higher than 200 cm/sq cm.

Table 1

Relative Importance of MCM Characteristics

The advantages of MCMs over PCSs may be summarized as follows:

• Higher chip density/reduced overall size

• Reduced mechanical and thermal stresses

(especially in the case of MCM-D)

• Lower interconnection delays

• Improved elearical performance

O High speed and frequency (more than

1 GHz)

• High power and voltage handling

• Increased reliability and system-level manufacturability

• High temperature capability with MCM-C and

MCM-D (up to 250°C)

O Limited only by solder melting

O Thermal design a critical issue

MCM Design Tools

At this time, the European consumption of

MC^s is approximately 25 percent of world production. It follows that a great deal of interest exists in MCM design tools. Such tools are currently offered by all broad-line EDA vendors, notably Valid Logic Systems Inc., Mentor

Graphics Corporation, Dazix, Racal-Redac Group

Limited, and Scientific Calculations. However,

Dataquest research indicates that a total of no more than 15 or 20 stations have yet been sold in Europe. Yet MCM tools are closer to systemlevel tools (which have heavy European consumption) than to IC tools, so why have so few been sold in Europe?

Sector

Automotive

Telecommunications

Consumer

Ojmputer

-PC

-Laptop/Notebook

-Workstation

-Mainframe

Mil/Aero

Source: Dataqoest (Decemtwr 1991)

Key MCM Interests

(In Order of Importance)

Cost, environmental, size, perfomiance

Cosi, f)erformance, size

Cost, size

CxDst, size, performance

Size, cost, perfonnance

Performance, cos:, size

Performance, size, cost

Environmental, size, performance, cost

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

i i i

CAP/CAM/CAE—Electronic Design Automation 17

Not surprisingly, MCM advantages depend on the perspective of the individual user company.

Dataquest research shows that the relative importance of MCM characteristics are heavily influenced by sector, as shown in Table 1.

At this time, MCM devices are relatively expensive. Hence, the high-end computer industry, where cost is relatively unimportant, is the major consumer. In Europe, a substantial amount of computer production, but relatively little computer design, takes place because non-

European computer companies manufacture equipment in Europe but design elsewhere.

However, a considerable amount of design is undertaken by the telecommunications, consumer, and automotive sectors, where cost is a key driver. This explains the moderately high

MCM consumption in Europe with relatively low penetration of MCM design tools.

For these reasons, we believe that the penetration of MCM design tools in Europe will occur in the following order:

• Military/Aerospace

• Computer

• Automotive

• Telecommunications

• Consumer

One of the basic problems with MCMs is that prototype devices cannot be tested and modified in the same way as PCBs. In other words, a prototype device is much more difficult to make design changes to using traditional cuttrack/wire-link methods. This means that more up-front analysis is needed in order to achieve a right-first-time design. In this sense, MCM design more closely resembles IC design.

Substrates for MCMs fit between PCBs and ICs.

ASIC layout tools would seem ideal for substrate design, except that MCM substrates are much larger, typically several inches per side.

Also, for IC design, 2 or 3 metal layers and

2 poly layers are the extent of IC interconnects, whereas MCM substrates and PCBs can have 20 or more layers. On the other hand, PCB design software, which is capable of large dimensions and many layers, is not typically suitable for track/gap dimensions of less than 3 mil.

Dataquest believes that M(3M technology will impact EDA tools in a number of ways. For example, the number of gates per gate array is currently rising, leading to problems of reduced yield, increased pin-out, and higher nonrecurring (NRE) charges. In the future, companies will bypass these problems by combining a number of lower-gate-count devices in a module. A very similar situation exists in the case of mixed-signal ASICs: Several lower-complexity (simpler) devices will be mounted in a mixed-signal MCM.

User Companies

We believe that the type of candidate company for MOI design tools will vary with MCM technology. Most systems companies involved in

ASIC design typically perform logic design and simulation, then pass a netlist to the ASIC manufacturer, which, in turn, performs placement and routing. This situation contrasts with that of PCBs, where the manufacturer typically undertakes no layout design. Hybrids are different again.

Before the late 1970s and early 1980s, most large systems companies owned a PCB fabrication facility. Over the years, these facilities have mosdy been sold off (often in management buy-outs) because of insufficient capacity utilization. However, those companies (especially in the automotive, aerospace, and military sectors) possessing hybrid manufacturing facilities have mosdy retained those facilities.

In the case of ASICs, systems companies very rarely manufacture their own devices, instead choosing to source devices from the merchant

ASIC marketplace. We believe, therefore, that

MCM-L and MCM-C layout designs will be lai;gely undertaken by systems manufacturers, while MCM-D layout will be undertaken by silicon MCM manufacturers. This situation is summarized in Table 2.

MCM design tools will not be driven by manufacturers in the same way that semiconductor manufacturers influenced the EDA market because the took are mainly a suite of new back-end tools designed to integrate with existing front ends. Also, PCB and hybrid manufacturers will undertake litde or no design and are, in any case, a fragmented group of relatively small companies possessing little individual power or influence when compared with semiconductor companies.

Product Requirements

An "ideal" MCM design station should provide facilities to allow the entire module to be designed using current top)-down methodologies,

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

18 CAD/CAM/CAE—Electronic Design Automation

Table 2

Layout Manufacturers b y Type

MCM Type

Manufacturer

MCM-L Specialized PCB manufacturer

MCM-C In-house or hybrid manufacturing company

MCM-D (Silicon) Specialized silicon MCM manufacturer

Source: Dataquest (December 1991) synthesizing individual devices if required. The building of a behavioral description of the entire m o d u l e should b e facilitated together with support for mixed-level simulation at the device, module, and system levek.

MCMs are characterized by several n a k e d dice mounted very closely in a confined space.

When this charaaeristic is coupled with another

MCM characteristic (high operating frequencies), several analysis problems result, as follows:

• Thermal issues—^Thermal problems are intensified through the p o p u l a r use of solderb u m p flip-chip device packaging a n d assembly because the major heat transfer path is through the solder b u m p s . Thermal analysis has never reached its potential as a system analysis tool. We believe that the technique is indispensable for MCM design, which will provide a welcome boost to the thermal analysis market.

• Parasitics—The extremely close proximity of signal-carrying tracks at high frequencies and the n e e d to provide impedance-matched s u b strates will inevitably require transmission-line analysis of crosstalk, reflection, a n d characteristic impedance. This n e e d will yet again raise the profile of companies, such as Quantic Laboratories, Q u a d Design Technology, and Swifdogic Limited, that supply this technology.

• Testing—Companies usually operate a multQevel testing strategy in connection with

MCMs: testing the chip, module, PCB, and system. Testing is a major problem at the module level because chips are densely packed into a sealed unit a n d also at t h e chip level because it is difficult t o test both bare dice and flip-chip packages. Yet individual chips must b e tested before assembly in order to minimize rework and repair.

MCM design will therefore bring into sharper focus t h e test problem, increasing the use of boundary scan testing a n d boosting the need for test logic synthesis.

CAE Design

Systems company

Systems company

Systeras company

Layout Design

Systems company

Systems company

MCM manufacturer

The expected increase in the u s e of lower gatecount devices within MCMs suggests a growing need for functional partitioning tools t o assist engineers in deciding w h i c h devices to assign to which modules. This n e e d is n o different fundamentally from the (currently unserved) need to partition at other levels (that is, at IC and system levels).

Another problem exists for MCM designers. The creation of dice outlines in parts libraries is difficult because semiconductor manufacturers d o not usually publish the data. Terminal dimensions are also a p r o b l e m for the same reason.

This often requires devices t o b e individually measured by MCM design a n d manufacturing companies—sometimes o n a per-batch basis! We expect systems companies to bring considerable pressure to bear on semiconductor companies in order to solve this problem.

Dataquest Perspective

The market for MCM design tools is still small.

However, many electronics systems companies are evaluating the technology a n d undertaking a n u m b e r of trial designs. Through this process, they h o p e to gain a competitive advantage by moving along the experience curve of this important technology in advance of other companies.

At this time, MCMs offer an expensive solution to packaging and interconnection problems.

Substantial European MCM design tools growth d e p e n d s on considerable price reductions in order t o be acceptable to consumer, telecommunications, and other major E u r o p e a n electronic sectors. Price reductions will only occur as a result of higher-scale economies.

Yet most existing M(^M sales are into high-end computer a n d aerospace/military applications, two declining sectors. So w h e r e are these scale economies to c o m e from? We believe that they will come from t h e high e n d of the other sectors, gradually creeping lower into (eventually) s o m e low-cost consumer p r o d u a s . However, a finite risk exists that this growth will never occur.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

CAP/CAM/CAE—Electronic Design Automation 19

Even so, MCMs are without a doubt one of the major packaging "events" of this decade, solving many problems and offering many advantages over PCB-only solutions. In partnership with PCBs, MCMs can achieve exceptional growth in the next five years, but design tools must keep pace with this technology. Renamed existing PCB or IC layout tools are not adequate for the MCM design task, and vendors that focus on the emerging segment of providing quality tools can expect to achieve significant growth in the next several years. •

By Jim Tully

believe that the latter point will be of major importance in post-1992 Europe in view of recent EC legislation.

SNI made significant advances to its EDA market position in Europe in the past two years— especially in Germany. However, a good proportion of this increase was because of captive sales into the Siemens parent company, which is unlikely to be repeated. Mentor is the market leader in Germany, which is Europe's biggest EDA market at $318 million in 1990 and forecast to have an 18 percent compound annual growth rate (CAGR) in the next five years. Dataquest believes that

Mentor's position in Germany will be enhanced by this agreement—but only marginally.

N e w s a n d V i e w s

Mentor Graphics and

Siemens Nixdorf

Enter Marketing

Partnership

Mentor Graphics Corporation of Wilsonville,

Oregon, and Siemens Nixdorf Informationssysteme (SNI), of Munich, Germany, recently signed a cooperative marketing agreement under which the companies will work together in

Europe. SNI has exited from the electronic design automation (EDA) marketplace, instead choosing to partner with Mentor Graphics in broad offerings for total solution sales throughout Europe (including Eastern Europe). The deal gives Mentor access to SNI's technology and distribution channels. Mentor will bring in

SNI as a partner for mechanical, electromechanical, and manufacturing applications.

This agreement follows the announcement of intention for SNI and Cadence Design Systems

Inc., of San Jose, California, to form a joint venture company in central Europe designed to boost Cadence's position in Europe and open distribution channels for SNI's produas and technology. This deal was killed by the recent

CadenceA^alid Logic Systems Inc. merger in view of Valid's already strong position in

Europe.

The deal also follows SNI's acquisition of Calay

Systems GmbH in 1990. We believe that the payback time scale on this acquisition, in view of additional heavy development expenditure, proved unacceptable for SNI (especially in view of the company's overall financial position).

SNI's original intention of becoming a broadline EDA vendor, which was largely fueled from internal development and acquisition, proved unrealistic, forcing the company to seek partners. Mentor Graphics should prove to be a valuable "customer" of SNI technology and a useful partner in SNI's system sales endeavors. •

Dataquest Perspective

From Mentor's perspeaive, the company gains access to key SNI technology in the fields of automatic test pattern generation and electromagnetic compatibility (EMC) analysis. We

By Jim Tully

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012516

20 CAD/CAM/CAE—Electronic Design Automation

In Future Issues

Watch for reports on the following topics in upcoming issues of Dataquest Perspective:

• An overview of the top-down design market

• A look at the logic synthesis market

• An analysis of hardware description languages

t

\

For More Information . , .

Managing Editor Patrick Waurzyniak (408) 437-8132

On the topics in this issue CAD/CAM/CAE (408) 437-8132

About other Dataquest publications Sales (408) 437-8000

About upcoming Dataquest conferences Conferences (408) 437-8245

About your subscription Customer Service (408) 437-8402

Via fax request Fax (408) 437-0292

The content of this report represents our interpretation and analysis of information generally available to the public or released by responsible individuals In the subject con^»nies, but is not guaranteed as to accuracy or completeness. It does not contain material provided to us in confidence by our clients. Individual companies reported on and analyzed by

Dataquest may be clients of this and/or other Dataquest services. This Information Is not furnished In connection with a sale or ofFisr to sell securities or in connection with the solicitation of an offer to buy securities. This firm and Its paient and/or their officeis, stockholders, or mend>ers of their ^mlUes may, from time to time, have a long or short position in the securities mentioned and may sell or buy such securities.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose,

0012516

CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

{

Dataoyest

n n acompamoi

MMMW The Dun STBradstrcct Corporation

Dataquest

Perspective

INFORMATION RESOURCE CENTER

DA rAQUEST INCORPORATED

1290 Ridder Park Drive

San Jose, CA 95131-2398

(408) 437-8600

CAD/CAM/CAE

Electronic Design Automation

Vol. 1, No. 2

December 2, 1991

Market Analysis

EDA in Europe: An Update

The European electronic design automation (EDA) market went through more than its share of peaks and valleys in 1990, as EDA companies in Europe experienced a combination of growth and retrenchment by some vendors while also seeing the rise of niche players and a diversity in growth patterns by various European countries. This research closely examines the forces at work in the European EDA market during 1990, a year in which

European EDA vendors were not spared the turmoil characterizing EDA markets elsewhere in the world.

By Jim Tully Page 2

The European Community's EMC Mystery Is Close to a Solution

As the dock ticks toward the unified European economy in 1992, several barriers exist in the approaches taken to Electromagnetic Compatibility (EMC) among European Community

(EC) countries. This research identifies some of the conflicts between EC countries over the current EMC disparities in member countries and offers an analysis of what must be done by electronic design companies to improve design tools for this nascent market.

By Jim Tully Page 8

News and Views

Viewlogic Systems Piles Initial Public Offering

By Ron Collett Page 11

©1991 Dataquest Incorporated / 1290 Eidder Park Drive,

San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

CAD/CAM/CAE—Electronic Design Automation

Market Analysis

EDA In Europe:

An Update

Introduction

Europe has not been spared in the electronic design automation (EDA) market tunnoil during the past two years. A combination of growth and retrenchment in specific application areas, the rise of niche players at the expense of major EDA companies, and different growth rates among European countries has created highly volatile market conditions. This article discusses and analyzes the following topics:

• European EDA market overview

• Vendor position

• Market drivers

• Differentiating factors in European EDA

B EDA case example—telecom industry

European EDA Market Review

In 1990, the European EDA software market was estimated at $307.4 million (see Table 1).

Hardware accounted for $463.9 million, and service revenue was a further $188 million.

Total revenue for Europe therefore totaled

$959.3 million. Preliminary analyses suggest that the market was flat in 1991, although selected areas such as logic synthesis and simulation appear to have experienced growth.

Growth of 13.3 percent overall was achieved in

1990. The computer-aided engineering (CAE) market grew at a very healthy rate of 32.3 percent as companies increased their investment in front-end productivity-enhancing tools. Most of this growth came from simulation tools, and some testing of the waters occurred in VHDL synthesis and simulation tools.

Table 1

European EDA Software Revenue and Growth

(Millions of Dollars)

CAE

PCB Layout

IC Layout

Total

Source: Dataquest (December 1991)

1989

135.9

108.1

27.4

271.3

Printed circuit board (PCB) layout revenue fell

9.3 percent after a healthy year in 1989. No major changes in device packaging or PCB manufacturing technology occurred in 1990, and

Dataquest's research showed that customers saw no need to invest further in PCB layout tools during this time frame. In other words, existing tools were good enough for the job.

The IC layout market was valued at $29.6 million in 1990, which is much smaller than the corresponding markets in the United States and

Japan. Moderate growth of 8 percent occurred in this market. "iTiis relatively small market size follows from the fact that most semiconductor companies have their major facilities in either

Japan or in the United States.

It makes sense that most standard IC components (such as microprocessors, memories, and standard logic parts) will be designed in those two countries. Europe's major usage of IC layout tools is in connection -with ASICs (including full-custom for consumer applications), not standard parts. Even so, the European ASIC market is smaller than the North American and Japanese markets. Together, these factors explain the relatively small size of the European IC layout market. However, when looking to the future, Dataquest believes that multinational semiconductor vendors will try to move closer to their end users in Europe and that this will increasingly lead to the design of standard parts in Europe.

Dataquest forecasts that the overall EDA market will continue to grow steadily through 1995 at a compound annual growth rate (CAGR) of

16 percent. This rate is higher than that achieved in 1989 and 1990, which is due mainly to a projected improvement in European economic conditions during the forecast period.

Vendor Position

Mentor Graphics Corporation remains the

EDA market leader in Europe, followed by

1990

179.8

98.1

29.6

307.4

Percentile Growth

32.3

-9.3

8.0

13.3

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

CAD/CAM—Electronic Design Automation

Racal-Redac Group Limited and Valid Logic Systems Inc. (see Table 2). All three companies faced problems in 1990 but fared well under the circumstances. Mentor is positioned well for

1992, given its Release 8.0 availability, shipments on the Sun platform, and its recent agreement with Siemens Nixdorf Informationssysteme AG.

Most of the larger European EDA vendors lost market share in 1990; only Cadence Design Systems Inc., Siemens Nixdorf, and Calay Systems

GmbH, which was acquired by Siemens Nixdorf during 1990, gained appreciable market share.

The other winners were niche CAE companies including Synopsys Inc., Viewlogic Systems Inc., and EEsof.

Country Position

Figure 1 shows the EDA market share position held by each of the major European countries.

Germany is currently the largest market, driven by its large economy, large industrial and automotive sectors, and buoyant consumer and data processing sectors. In 1990, the United

Kingdom lost its second place standing to

France, with its important consumer electronics, automotive, teleconmiunications, and defense sectors. Third position is held by the United

Kingdom with its strong (and deregulated) telecom market.

From a future growth viewpoint, Dataquest forecasts that Germany will again lead the EDA market, followed by Italy, Scandinavia, France, and Benelux. This change in ranking order is mainly driven by the relatively poor economic conditions in the United Kingdom and reduced defense spending in France. Additionally, we expect Itdy to attraa substantial Japanese investment in the next three to four years, being geographically located at the opposite end of Europe from the United Kingdom

(which to date has attracted the lion's share of

Japanese investment in Europe). Italy is attractive because of its geographical position, good facilities, and skilled work force.

European EDA Market Drivers

The following key drivers currently influence the European EDA market.

Distribution and Support

Because Europe cunentiy is, and will continue to be long after 1992, a group of separate countries, each with its own language and culture, both distribution and support are key issues. For example, it is not possible, or even acceptable, to emulate a typical U.S. company situation and install a single toll-free 8(X) telephone number as a support hotline for all of

Europe. Support must be dose to the customer

(in other words, usually within the same country) and staffed by personnel who are not only

Table 2

Top 10 European EDA Vendors in 1990—Software Revenue and Maiiiet Share

Company

Mentor Graphics

Racal-Redac

Valid Logic

Cadence

Dazix

Siemens Nixdorf

Computervision

Calay

EEsof

Ziegler

Others

Total

NA = Not available

Source: Dataquest (December 1991)

MJllions of Dollars

1990

50.2

32.2

23.8

23.3

18.0

11.0

8.3

8.3

6.8

6.6

118.9

307.4

1990

Market Share (%)

1989

16.3

18.8

10.5

7.8

13.9

8.6

7.6

5.8

3.6

2.7

2.7

2.2

3.1

8.2

0.2

2.2

38.6

100.0

2.8

2.5

NA

2.0

39.9

100.0

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

CAD/CAM—Electronic Design Automation

Figure 1

Country Positions—Software Revenue

Germany

France

United Kingdom

Italy

\^:sst^^^<:^^^^s>x<^^^^^

K^^^^xsx^^^<^^w^^x^

^sssss^c^^sssssss^

:'<\XV%XM

Scandinavia

Benelux

Spain

.X%X?'.V\1

<SSSSS1

Srvj

Rest of Europe

S^^^SS^4

( i 20 40 60 ao 100 120 140

Millions of Dollars

Source: Dataquest (December 1991) technically proficient but also fluent in the local language. It follows that European EDA vendor organizations can be expected to be more expensive to operate than U.S.-based operations of equivalent size.

Economic Conditions

Clearly, economic conditions will impaa enduser industries and have a corresponding effea on EDA sales. The market dynamics during economic downturns are, however, complex.

The major forces currently acting on the market include the following:

• End-user companies' revenue and income is reduced, leading to corresponding reductions in EDA tool purchases.

• Under pressure to reduce head count, some end-user companies may turn to greater levels of design automation, thereby increasing

EDA spending while reducing head count.

The economics of such a decision will vary among countries. For example, in France and

Germany, companies are subjea to heavy financial overheads on the employment of individuals.

• Some engineers and design draftsmen who are laid off during a downturn start small design services companies and purchase lowend EDA tools using layoff compensation or other funds. In some cases, this scenario boosts PC-based EDA sales.

When considering economic conditions, it is always necessary to look at European countries individually. For example, although Germany's economy is very large, it will experience slower growth in the near term largely because of the reunification of East and West (Germany.

Meanwhile, the United Kingdom is rising out of recession and therefore will exhibit some signs of growth. Even so, the German economy will remain the biggest in Europe, while the United

Kingdom's economy is still delicate.

Industry Consolidation

Consolidation of end-user industries is a very important faaor in the European EDA industry, because it directly impacts the supply/demand balance. The European electronics industry has been restructuring during the past several years.

This is in preparation for the single market of the unified European Community (EC) of 1992 and for reasons related to industry evolution and maturity. Virtually every major European electronics company is involved. This consolidation raises a number of very important and fundamental issues affecting the balance of power between vendor and customer, and it also raises the question of which vendors' tools will be chosen after a merger or acquisition. All vendors are affected by this situation.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

CAD/CAM—Electronic Design Automation

Eastern Europe

The effects of the changes in Eastern Europe are shown as follows:

• Eastern European countries require major infrastructure development. One of the first areas that need to be addressed is telecommunications. Western European companies can expect to receive the lion's share of this business because of their historical trading and cultural ties and because of the importance of Western European companies in the worldwide telecommunications market. This will boost European EDA sales.

• Eastern European electronics companies will gradually begin to purchase EDA systems.

However, Eastern European economic growth will continue to be low, and Eastern European companies will have difficulty paying in hard currency. Dataquest does not expect this market to show signs of meaningful growth until the latter half of this decade.

Differentiating Factors in European EDA

From the perspective of electronic systems marketing, one of the most obvious differences between Europe and the rest of the developed world is its demographic diversity. This is summarized as follows:

• Western Europe—2A countries, 370 million people

• Eastern Europe—8 countries, 395 million people (with the USSR as one country)

• Languages—^25 (minimum)

Clearly, Europe is not a single entity but a complex, multicountry environment with a myriad of cultures, languages, and styles of doing business. These considerations must radically shape any marketing strategy targeted at

Europe. It is also one of the biggest markets in the world for many types of electronic products. In 1991, Europe will become the largest CAD/CAM/CAE market in the world, taking over this position for the first time from

North America.

The 1992 Single Market

A key factor that will radically shape the competitive scene in Europe is the Single European

Act in 1992. This act will result in a single

European market of 320 million people.

One of the Act's major global effects will be to encourage companies currendy exporting to the

EC to set up manufacturing plants in Europe in order to bypass quotas. Indeed, this has been happening over the past several years in preparation for 1992. However, we strongly believe that these manufacturing plants will have to add design capabilities over a period of time. It takes too long to communicate change requirements from design centers in Europe to design centers in either Japan or the United States, then have new manufacturing instructions fed back to manufacturing plants in Europe.

The Single European Act does not require that design be undertaken within the EC, so the direct, short-term effea on EDA revenue will be minor. The main electronics design implication of the Act relates to standardization of product safety, electromagnetic compatibility (EMC), and other related regulations. Some additional growth can be expected because of the need to redesign certain products to comply with standardization regulations.

An EDA Case Example:

Teleconununications

w h e n searching for EDA opportunities, it is useful to segment end-user companies into groups that exhibit broadly similar needs. In view of the importance of the telecom sector to

European EDA revenue, the specific needs of this industry are examined here more closely.

Figure 2

Electronics Industries In Europe—^1990

End-User Revenue

j^ \ Industrial / ^ ^

/ \ ^^•''^ /Milltary/X

/ \ /AGrospaceX

/ \ / 1 1 , 7 % ^ ^

/ Telecom/ \ / ^.^^''^

/ComrnunicationsX / ^^--''''^

\ Transportation

1 6 . 2 %

IQ fl«, V - ^

\ Consumer

\ 21.9%

Data

Processing

25.8% J

Total Revenue = $191,000 Million

Source: Dataquest (December 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

CAD/CAM—Electronic Design Automation

Europe's most important electronics industries are shown in Figure 2. They are ranked in order of production revenue. Europe is more evenly segmented than the United States or

Japanese markets, with all of these industries receiving a reasonable slice of the pie. Data processing is the largest sector with $54 billion, followed by the consumer electronics market.

Mixed-Signal and Analog

Design Tools

Mixed-signal and analog design tools are key issues for both ASIC and system-level design.

The market for mixed-signal ASICs in Europe grew 68 percent in 1990, representing one of the highest growth areas in the ASIC market.

The leading European mixed-signal ASIC supplier, Mietec, grew its revenue primarily through its involvement in the telecom market— specifically, its involvement with Alcatel's System

12 exchanges.

However, many data processing and consumer electronics manufacturing companies do not design products in Europe. Such companies therefore are not potential customers for EDA tools in Europe (yet). The picture changes radically when considering investment in electronic design (see Figure 3). In this case, the ranking changes, with telecom in the lead by a considerable margin. Figure 3 also shows each sector's percentage of total electronic design investment. This investment includes expenditures such as design tools, personnel, and buildings.

Europe's importance as a telecom center is further illustrated by noting that European companies occupy 5 of the top 10 positions in the worldwide telecom equipment market share rankings.

EDA Needs a n d Opportunities i n Telecom

Certain specific needs of telecommunications companies, as shown by Dataquest research, are described in the following paragraphs.

Figure 3

Investment in Electronic Design In Europe, by Sector (1990)

Transportation

5.1%

A recent Dataquest survey of European ASIC design centers showed that the lack of adequate mixed-signal design tools is the greatest cause of concern to ASIC designers.

The lack of analog synthesis tools and insuffidendy integrated analog and digital simulation tools stand out as specific issues of concern.

Analog PCB layout is another shortfall area.

Many telecommunications companies involved in analog design have long since abandoned any attempts to use auto-routing tools and are forced to use manual techniques. The market is ready and waiting for these tools.

High Clock Frequency Layout and

Transmission Line Analysis Tools

High dock frequendes are a characteristic of the telecom industry. This is especially true for central exchange and RF digital applications, which leads to specific PCB layout problems and a need for die following:

• Emitter-coupled logic (EO.) balanced-load layout

• Noise-redudng power plane design

• Minimization of reflection and crosstalk, using appropriate analysis tools

Most vendors have now recognized these needs, which accounts for the recent intense activity from major vendors in this area, espedally from Racal-Redac and Valid Logic.

Telecommunications equipment is often processing noisy signak in harsh environments. An ability to analyze and predict noise levels is therefore key. Most telecom companies look for accuracy in these tools in preference to speed of operation.

Total = $8,297 Million

Thermal Analysis Tools

Thennal problems are closely assodated with increases in system frequendes. Because

Source: Dataquest (December 1991)

©1991 Dataquest Incorporated / 1290 Eidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

CAD/CAM—Electronic Design Automation extreme temperature is a major killer of silicon devices, the existence of tools to better predict the thermal operation of systems is attractive.

Most major EDA vendors now have a thermal analysis capability. Mentor Graphics, Racal-

Redac, Cadence, and Valid Logic all possess in-house thermal analysis tools. These vendors consider such tools to be strategically important and have invested considerable sums in company acquisitions, technology acquisitions, or internal development to obtain them.

Dazix currently integrates the Pacific Numerix product.

Telecom equipment is often of multiboard rackmounted construction. A strong interest therefore exists in simulating thermal performance at the overall enclosure level. Very few products exist that can simulate thermal effects at the system level. One company operating in this field is Flomerics limited in the United

Kingdom.

DSP Design Tools

Dataquest predicts that the highest growth areas in the telecommunications field during the next 10 years will be video conferencing, video telephony, and multimedia applications.

All require sophisticated data compression techniques in order to be viable. European telecom companies are deeply involved with these technologies and are therefore intensely interested in standards such as Joint Photographic Experts Group, or JPEG (for still picture compression) and Motion Photographic

Experts Group, or MPEG (for moving picture compression).

Audio compression for voice wireless communication is another hot issue. Digital signal processing (DSP) techniques are essential for these applications. Only two vendors have market visibility in this area—Mentor Graphics

(via its joint venture with NV Philips in Belgium) and Comdisco Systems Inc. These applications are also generating interest from semiconductor companies that offer applicationspecific standard products (ASSP) and ASIC products. Such companies include LSI Logic

Corporation, Texas Instruments Inc., and VLSI

Technology Inc.

PLD and FPGA Technologies

Programmable logic devices (PLDs) and fieldprogrammable gate arrays (FPGAs) will be needed for both prototype and production purposes in the telecom sector. However, we believe that they will be used mosdy for production purposes because of the limited production runs and high degree of customer tailoring associated with high-value telecommunications equipment.

EMI Analysis Tools

The interest in these tools is driven by the EC directive on electromagnetic compatibility: designated 89/336/EEC (see related research, "The

European Community's EMC Mystery Is Close to a Solution," on page 8). This applies to many sectors, not only telecom—especially computers and consumer electronics products—but telecom equipment has particularly stringent requirements in exchanges, in the home, and for portable applications.

The standard sets limits to the maximum allowable electromagnetic radiation transmitted from electronic equipment. It also controls the ability of equipment to withstand incoming radiation.

This standard is part of the 1992 standardization efforts, and it is prompting an interest in electromagnetic interference (EMI) analysis tools, which will simulate and predict field strength a specified distance from a piece of electronic equipment. Such tools are not commonly used or commonly available from vendors at this time. Siemens Nixdorf and Swiftlogic Limited currently operate in this area.

Dataquest Perspective

Although the European EDA market achieved moderately good growth in 1990 despite relatively poor economic conditions, 1991 appears to have been somewhat disappointing.

Tliis regional market shares many characteristics with the North American and Asian markets, although significant differences exist such as in the telecommunications arena.

Dataquest believes that companies that search for and successfully address these differences will achieve a competitive advantage. Furthermore, this type of differentiation strategy is one of the few ways in which smaller niche companies can enter the market and begin to penetrate in a meaningful way. •

By Jim Tully

©1991 Dataquest Incorporated / 1290 Ridder Parle Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

8 CAD/CAM—Electronic Design Automation

The European Community's EMC Mystery Is

Close to a Solution

The European Community (EC) aims to create a single market that ensures the free movement of goods and services between member countries. But many subtle barriers to the free movement of goods currently exist, requiring legislation for their removal. One such barrier is the varying approaches to electromagnetic compatibility (EMC) among countries. This article examines the concept of EMC, the need for

EMC testing, its impact on EC trade, and the current (and unserved) demand for EMC simulation tools. unwelcome electromagnetic radiation. Such problems will multiply with increases in system dock speed and operating frequencies. Many circuits already operate at ftequendes where

EMC problems are apparent, yet microprocessor dock frequendes are forecast to rise 15-fold during the next decade (see Table 1).

As a result. Directive 89/336/EEC was issued

May 3, 1989, and ofBdaUy becomes effeaive on

January 1, 1992 (although in reality, a transition period of several years is likely to be agreed upon). The Directive sets limits on maximumallowable radiated emissions, conduaed emissions, susceptibility to power variations, common-mode radio frequency interference, electrostatic dischaige, and radio frequency fields.

Background

Most EC member countries legislate for maximum permissible electromagnetic disturbance generated by electronic equipment. The problem is that many of these regulations are incompatible and constitute a hindrance to trade within the EC. Although a piece of equipment may comply with one of the more stringent country regulations (for example, Germany's or VDE regulations), it may fail to be accepted in another country because it does not bear the mark of approval for that specific country. TTiis demonstrates one of the subtle barriers to trade that cannot be tolerated in post-1992 Europe.

EMC may be a hot political issue in Europe, but there also are sound engineering reasons why it will become stiU more important. Random errors within computer systems and annoying interference to TV reception are only two examples demonstrating the effects of

GaAs (Logic/ASIC)

Source: Dataquest (December 1991)

The Directive legislates for equipment's effect on the immediate environment and on the equipment's ability to operate trouble-free during incoming interference. Currently, all apparatus is subject to the Directive except motor vehide spark ignition systems, certain electridty tariff meters, and amateur radio apparatus

(which are the subject of other Directives containing EMC requirements). Apparatus that meet the requirements will bear the mark of approval shown in Figure 1.

Table 1

Forecast System Clock Speed Increases—1990-2000

Parameter

MPU Speed

Current

20 MHz

Near Term

1993-1995

100 MHz

Memory

Speed

CMOS

20-80ns

BiCMOS

9-60ns

ECL (Logic/ASIO

150^00ps 50-150ps

50-80ns

Implications on Design

Directive 89/336/EEC has generated intense interest throughout Europe and elsewhere on the subject of EMC. However, although many engineers are familiar with the design practices necessary to minimize electromagnetic interference (EMI), the add test comes at the conformance testing stage after a prototype has been submitted for EMC conformance testing. Such

20-60ps

Long Term

1997-2000

300MH2

BiOlOS/FEERAM

<25os

1.5-2.0ps

(yhotonic Lc^c)

<5-10ps?

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

{

K i

CAD/CAM—Electronic Design Automation

Figure 1

EC Conformity Mark

Note: This mark will appear with the year in which it was affixed.

Source: European Community

tests consist of directing an antenna at the equipment under test, at a prescribed distance, while sweeping all frequencies with a spectrum analyzer. Another test analyzes energy transmitted along the power cord. Testing time varies from approximately two days for simple produas to two weeks for complex products.

The foEowing two implications arise from a produa design viewpoint:

• Time to market—Dataquest research has shown that most conformance tests fail at the first attempt, prompting a further design iteration in order to correa the problem. This leads to healthy business for the testing houses but increases products' time to market because of the time involved in additional design iterations. In today's competitive environment, this issue is very serious.

• Product cost—In many cases, electronic products will require additional filtering or screening in order to comply with the legislation. This is most serious in high-volume, low-cost product sectors such as consumer electronics. These sectors will be prepared to invest considerable sums in minimizing EMC effects within the electronic design in order to minimize filtering and screening costs.

Clearly, both of these issues affect all electronic systems companies, but Dataquest believes that small companies will be particularly badly affected because most do not have the resources to devote to EMC issues. German companies should fare better than most because they have been required for some time to comply with EMC regulations (originating with the

German VDE standards body), which are at least as stringent as 89/336/EEC.

EMC Simulation Tools

In view of companies' experience of EDA tools, engineers are increasingly asking "Why can't we simulate electromagnetic interference effects before bxiilding a prototype?" This would minimize the number of design iterations, but very few tools are currently available for this purpose. We believe this is because of the following reasons:

• The technical problems of producing EMC simulation tools, based on highly mathematical electric field theory, are considerable. A very small number of vendors have developed or acquired the necessary technical expertise to produce such tools.

• Most EDA tools users operate at the IC or

PCB level, but EMC analysis cannot be totally carried out at this level because the key measure is field strength outside the enclosure. This means that engineers of many disciplines (including electronic, mechanical, thermal, electrical, and materials) are involved. Individual requirements are therefore fragmented across these disciplines.

A key area of difficulty in analyzing EMC is to identify and locate the source of the emission problem; it is insufficient to simply report that a problem exists. The problem is multiplied because the symptoms of the problem are usually separated from the root cause. For example, radiation can often be traced to the cables that interconnect subsystems, yet the circuitry on PCBs is usually ultimately responsible for the emissions—and this is where the problem must be fixed.

Problems can be purely electrical (edge transition times being too fast) or physical (bad practices in PCB layout such as ground loop construction rather than star or ground plane construction). In order to analyze the sources of these problems, the simulator must have knowledge of the physical parameters (in three dimensions) and materials used. These points illustrate the difficulty of providing a viable

EMC simulator.

The Ideal Product

Early EMC simulation products will operate at the PCB level. These products will partially satisfy user requirements, but a quantum leap in functionality is needed in order to fully cater to users' needs.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

10

CAD/CAM—Electronic Design Automation

Dataquest's end-user research shows that the solution to minimizing emissions lies partly with recognized design practices in terms of layout, line lengths, edge times, and so on, and partly with a design procedure more closely resembling a "black art." On this case, black art refers to designs learned through experience rather than academic studies.)

At this phase of the design, engineers ideally need "what-if' analysis tools, which will simulate, in real time, the effects of the following examples:

• Edge transition times (perhaps via the inclusion of series resistors to reduce the speed of a transition)

• Different clock frequencies

• Cable length and position variations

• Various enclosure coatings

The simulator should identify regions of high field intensity in three-dimensional space around the simulated equipment according to restrictions that correspond to a particular EMC standard (such as 89/336/EEC). The frequency characteristics and other parameters should be reported to the user. Directional vectors should be calculated and automatically used by the simulator to locate the sources of the emission.

The engineer must then dedde whether to undertake further tracing of the root cause or employ shielding techniques around emission locations.

The Vendors

Right now, a demand for tools exists, but no vendors are yet shipping products. Therefore, the EMC simulation market does not yet exist.

However, vendors are beginning to take a keen interest in the topic, and we expect several companies to introduce products within the next year. Companies that are active in this field include Swiftlogic Limited (Cumbernauld,

United Kingdom), Quantic Laboratories (Winnipeg, Canada), Siemens Nixdorf Informationssysteme (Munich, Germany), and Quad Design

Technology (Camarillo, California). These companies are all active in the transmission line analysis sector, which is mathematically related to EMC simulation.

One of the main problenas faced by EMC simulation tools is the performance/accuracy tradeoff. Users ideally want instant (real-time) analysis to an accuracy -within 5 to 10 percent.

However, we believe that this degree of accuracy will require considerable processing time and that more acceptable tools will emerge in the 1993 to 1994 time frame following considerable increases in -workstation performance and more efficient algorithm development.

Market Size

Ln assessing the size of this not-yet-existent market, it is useful to consider the value companies would place on such tools. Dataquest's research has shown that, in Europe, for companies new to EMC, approximately 75 percent of products submitted for conformance testing fail at the first attempt For experienced companies,

30 percent fail at the first attempt. Overall, approximately 40 percent of all conformance tests fail at the first attempt. Relating these figures to the product cost and time to market issues discussed previously, we believe that a seat price of $60,0(X) to $80,000 can be commanded, which in the first year translates to a

European market valued at approximately

$12 million.

Dataquest Perspective

The EMC simulation market, triggered by EC legislation, is ready to take off in Europe. Only product nonavailability is inhibiting this process.

Tool technology is exceptionally difficult to develop, but high rewards await those vendors that ultimately offer workable solutions (in

Europe and elsewhere). The whole area of EMC is shrouded in mystery and is understood by a small number of experts using manual "rule-ofthumb" techniques. Dataquest believes that, spearheaded by a handful of innovative vendors, the mystery will soon to be solved, opening the field of EMC analysis to the mainstream electronic engineering community. •

By Jim Tully

(This research was also published by

Dataquest's European Semiconductor group.)

©1991 Daaquest Incorporated / 1290 Bidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

G\P/CAM—Electronic Design Automation

11

News and Views

Viewlogic Systems Files

Initial Public Offering

Viewlogic Systems Inc. last month filed a registration statement with the Securities and

Exchange Commission for an initial public offering (IPO) of 3 million shares of common stock, of which 1.8 million will be sold by Viewlogic and 1.2 million will be sold by certain stockholders.

The currently estimated IPO price will be between $10 and $12 per share. Viewlogic, of

Marlboro, Massachusetts, intends to use the proceeds from the offering for working capital and general corporate purposes. The shares will be offered through an underwriting group led by

Donaldson, Lufkin & Jenrette Securities Corp.,

Alex. Brown & Sons Inc., and Wessels, Arnold

& Henderson.

Viewlogic, which was established in October

1984 and shipped its first commercial products of its Workview software in July 1985, first became profitable in 1989 when the company earned a net income of $646,000, or 7 cents per share, on revenue of $17.2 million (see

Table 1). For the year ended Dec. 31, 1990,

Viewlogic showed net income of $2.7 million, or 27 cents per share, while nearly doubling its revenue to $30.4 million. For the nine months ended Sept. 30, Viewlogic's earnings were

$1.98 million, or 19 cents per share, on revenue of $29.8 million.

The CAE software segment, on which Viewlogic has concentrated its energies, is the largest and one of the fastest-growing segments of the EDA market. Dataquest estimates that total CAE software sales will reach approximately $1.3 billion by 1994.

Dataquest Perspective

With this public offering, Viewlogic joins a host of other EDA-only companies that are public, a list that includes Mentor Graphics Corporation,

Cadence Design Systems Inc., Valid Logic Systems Inc., IKOS Systems Inc., Zycad Inc., and

Silvar-Iisco. Dataquest believes that Viewlogic needs the capital in order to compete effectively against the larger EDA companies, especially in Asia, which will represent a significant

EDA opportunity during the next three years.

The revenue generated by the offering, which would equal a minimum of $30 million assuming the low end of the estimated IPO per share price of $10, should give Viewlogic a solid financial position. Clearly, it will allow the company to increase its investment in R&D, sales, and marketing.

By Ron Collett

Table 1

Viewlogic Systems Inc.

Summary Consolidated Financial Information

(In Thousands of Dollars, except Per Share Data)

Statements of Operations Data

Revenue

Income (loss) from operations

Income Ooss) before extraordinary item (1)

Net income Ooss)

Pro forma per share data (2)

Income Ooss) before extraordinary item

Net income Ooss)

Pro forma weighted average number of shares

Year Ended December 31

Nine Months

Ended

September 30

1986 1987 1988 1989 1990

1990 1991

3,581 6,688 9,898 17,170 30,380

20,798

29,755

-2,112 -748 -^35 565 2,726

2,066

2,916

-2,065 -832

-542 407 1,413

1,084

-2,065 -832

-542 646 2,735

1,987

1,582

1,582

-0.38 -0.13 -0.07 0.05 0.14

0.10

-0.38 -0.13 -0.07 0.07 0.27

0.19

5,489 6,282 7,845 8,636 10,284

10,218

0.15

0.15

10,774

Source: Viewlogic

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

12

In Future Issues:

Watch for the following topics in upcoming issues of Dataquest Perspective:

• An overview of the Top-Down Design Market

• IC Layout Market Update

CAD/CAM—Electronic Design Automation

i i

For More Information

Managing Editor Patrick Waurzyniak (408) 437-8132

On the topics in this issue CAD/CAM/CAE (408) 437-8132

About other Dataquest publications Sales (408) 437-8000

About upcoming Dataquest conferences Conferences (408) 437-8245

About your subscription Customer Service (408) 437-8402

Via fax request Fax (408) 437-0292

The content of ttils report represents our interpretation and analysis of infonnation genetally available xa the public or released by responsible iodMcbials In the subject con^nuiies, but is not guaranteed as xo accuracy or completeness. It does not contain material provided to us in confidence by our clients. Indhridual companies reponed on and analyzed by

Dataquest may be clients of this and/or other Dataquest services. This infonnation is not furnished In connection with a sale or oHer to sell securities or in connection with the solicitaiion of an offer to buy securities. This finn and its parem and/or their officers, stockholdets. or meiribers of their fainilies may, ftom time to time, have a long or shon position in the securities mentioned and may sell or buy such securities.

©1991 Dataquest Incoipoiated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0012398

I

INFORMATION RESOURCE CENTER

DATAQUEST iNCORPORATED

Dataqyest 1290 Bidder Park Drive

BB?ffSg£''.gLis.^corpo™.»n Sfi" J"=^e, CA 95131-2398

(408) 437 8600

Dataquest cm>umu

Perspective

Electronic Design Automation

Vol. 1, No. 1 November 11, 1991

Market Analysis

Next-Generation EDA: Electronic Systems Design Automation

Today's electronic design marketplace poses a plethora of challenges to electronics manufacturers: global competition, shrinking product life cycles, collapsing market windows, and increasing design complexity. Dataquest's research examines how designers will seek to maintain a competitive advantage by adopting new design technologies that integrate advanced electronic design automation OEDA) with other design automation disciplines. This emerging breed of technology will usher in the next generation of design automation— electronic systems, design automation (ESDA). Among the first ESDA areas that the market will demand will be the combination of high-level EDA tools with computer-aided software engineering.

By Ron Collett and John DeArmon Page 2

Interpreting the Mixed Signals of the Mixed-Signal Market

Gaining a firm grip on the mixed-signal design market has been a tall order for both chip and tools suppliers. The mixed-signal design market, which includes electronic systems and

ICs comprising a mixture of analog and digital circuits, continues to elude chip and EDA vendors. For example, many designers have foregone the mixed-signal ASIC approach to avoid its inherent high costs, high risks, long development cycles, and testing problems.

Dataquest's research shows, however, that although the mixed-signal EDA market has been a casualty of the market's aversion to designing complex mixed-signal circuits, strong signs show that the situation is about to change.

By Ron Collett Page 7

The New Business Model of the 1990s for EDA Vendors

To compete in the hotly contested EDA market of the future, large EDA software vendors must begin to consider fundamental changes in their current business models. In short, they must look toward outside sources for design automation products. Whereas EDA vendors have primarily relied on the in-house development of products, the most successful large EDA vendors will be those that provide a ridi combination of internally developed products and third-paity tools, Dataquest's research looks at the growing need for a robust

EDA OEM channel for sales of third-party technology and tools.

By Ron Collett Page 13

©1991 Dataquest Incoiporated / 1290 Eidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

CAD/CAM/CAE—Electronic Design Automation

Market Analysis

Next-Generation EDA:

Electnmic Systems

Design Automation

In today's electronic design arena, electronics manufacturers face a myriad of challenges: global competition, shrinking product life cycles, collapsing market windows, and shorter product design cycles. At the same time, design complexity continues to rise. Dataquest's research indicates that designers are looking to retain their competitive advantage by adopting new design methodologies together with more advanced design automation technologies. The goal is to more effectively automate the entire electronic systems design process.

With this as a backdrop, the stage is being set for the next-generation EDA opportunity— electronic systems design automation (ESDA).

ESDA merges different design disciplines, technologies, and methodologies into a tightly integrated development environment. Dataquest believes that ESDA tools will become a significant requirement for the electronic product development market beginning in 1993 and continuing throughout the decade.

High-Level, Top-Down Design +

Automated Concurrent

Engineering = ESDA

Dataquest defines ESDA as the combination of high-level, top-down design and automated concurrent engineering. It is the meshing of these two methodologies, together with the underlying design automation technologies supporting them, that forms the basis of our definition. Electronic designs demanding the use of ESDA tools include those having high complexity (in terms of hardware and/or software), size, and/or performance that are bounded by a range of rigorous constraints (for example, time to market, form factor, cost, power consumption, adaptability, and reliability).

The Institute for Defense Analyses (IDA) defines concurrent engineering as follows: "Concurrent engineering is a systematic approach to the integrated, concurrent design of products and their related processes, including manufacture and support. This approach is intended to cause die developers, from the outset, to consider aU elements of the product life cycle from concept through disposal, including quality, cost, schedule, and user requirements."

Concurrent engineering involves the continuous and incremental identification of design constraints across domains (for example, electronic design, software, and packaging) and shifting these design constraints to the front end of the design cycle while using those constraints to drive product development throughout the design cycle. This method contrasts with the way most design is done today, whereby actual design constraints, either inherent or imposed by other design domains, are discovered downstream in the design cycle often after a prototype is completed. The most severe design constraints latendy imposed today arise from software, hardware, and packaging incompatibilities. Concurrent engineering takes into account not only the design and manufacture of a complete system but also the ongoing field support

(that is, maintainability). In summary, concurrent engineering addresses both the product development phase and the subsequent life cycle of a product.

Vendors providing ESDA products that both shorten the development cycle and facilitate product designs yielding maximum value at the lowest cost will hold the potential to reap great rewards in the 1990s.

High-level, top-down design enables the designer to quickly specify all facets of an electronic product design's behavior at a high level of abstraction (for example, signal-flow diagrams, high-level languages, object-oriented design components, state transition, and dataflow diagrams) and rely on an integrated suite of hardware and software tools to synthesize and verify the ccoresponding physical implementation. Today's logic synthesis tools have begun to move the hardware designer into the realm of high-level, top-down design, but the technology is still in its infancy. The goal of highlevel, top-down design is to provide engineers with tools that enable them to focus on the total system architecture and behavior as opposed to only function and structure.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

CAD/CAM/CAE—Electronic Design Automation

ESDA currently has many gray areas. Generally, however, vendors providing ESDA products that both shorten the development cycle and facilitate produa designs yielding maximum value at the lowest cost will hold the potential to reap great rewards in the 1990s. It is clear that the merging of EDA and software development tools will be among the first ESDA market requirements. Our belief is based on research that shows an increasing emphasis on software development for next-generation products. Currently, the market devotes an average of 56 and

44 percent of its development resources on hardware and software development, respectively. Figure 1 shows that this balance wiU shift during the next two years, with software consuming an average of 51 percent of the development resources and hardware consuming

49 percent of the resources. Figure 2 illustrates the corresponding distribution of hardware and software resource allocation across the electronic design market. As shown, the largest single segment of the market allocates approximately the same amount of its resources to hardware and soflAvare development (41 to 50 percent).

Dataquest believes that the military and aerospace industries, as ivell as most major computer manufacturers, mil be at the forefront in adopting ESDA.

design engineer. Figure 3 also illustrates that in

1995, chip manufacturers will be able to integrate approximately 92 transistors in the same amount of area required for a single transistor in 1980. Dataquest believes that the widening gap between what can be manufactured versus what can be designed will be filled to a large extent by expanding the design's software content as well as by advancements in hardware design automation tools and increasing the reuse of existing designs.

ESDA is in its conceptual stage. Nonetheless, early arrivals setting their sights on this market include companies such as upstart Redwood

Design Automation Inc., Scientific and Engineering Software Inc., Ascent Logic Corporation, and

Comdisco Systems Inc., as well as a range of logic synthesis suppliers.

The Evolution of EDA and

Software Development Tools

Figures 4 and 5 depict the past evolution and projected future of the EDA and computer-aided software engineering (CASE) industries, respectively. In the 1970s, CAD/CAM companies such as Applicon, Calma, Computervision, and Scientific Calculations Inc. focused on automating

IC layout and/or printed circuit board (PCB) layout. At the same time, computer manufacturers were dominating the software development tool industry, although early software tools suppliers such as Lahey Computer, Ryan

McFarland, and Xerox PARC began creating fourth-generation languages, advanced compilers, and related low-level development tools in the

1970s. Incremental advances in both EDA and

CASE technologies continued throughout the decade.

Dataquest believes that the shift favoring an increase in allocation of resources for software development throughout the decade will be driven by the fact that semiconduaor manufacturing processes will outstrip hardware engineering design productivity. Advancements in semiconductor fiibrication processes will enable designers to put five times more transistors than is currently possible on a single chip by 1995.

Figure 3 shows the potential gap that will exist between improvements made in design productivity and device integration. The magnitude of design and manufacturing advancements are compared with what was feasible in 1980. Comparing 1980 with 1990, the EDA tools have enabled only a 5-fold increase in design productivity, whereas chip manufacturing has improved

23-fold. Thus, in the same amount of time, engineers are now able to design circuits that have five times more transistors. Currently, leading-edge design teams can implement approximately 4,000 gates per month per

Maintaining compatibility uHth the current generation of design automation tools will be a key ingredient for success in the impending

ESDA market.

EDA and CASE took a significant leap forward during the 1980s. Companies including Daisy

Systems Corporation, Mentor Graphics Corporation, Valid Logic Systems Inc., Viewlogic Systems Inc., Gateway Design Automation (now part of Cadence), HHB Systems (now part of

Racal-Redac Inc.), and Synopsys Inc. entered

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

CAD/CAM/CAE—Electronic Design Automation

Figure 1

Average Percentage Investment in Developing Hardware Portion of Systems Design versus

Software Portion of Systems Design

Percentage

^ ' ' " ' f f ^ ^ ^ '—$^^^»g^i^^^

90

Average Percentage of Resources Invested in Software Development j-j^h^. J * - M ^ T S J I

^ ^ " ^ • ^ • ^ ^ S ^ ^ - ' ^ V ^ ^ ^ i S ; : ; " ; - -v-^ ? ^ : ^ r - ^ ^ i ^ - ^ ^ ^ 1 : ' ^ r 4 ^ ^ V ^ ^ f r x . * - ; ^ ^

yo-i-.eri

60

^^^^mS^i^'^^^'^W^-.

304

WM^m^WZ/f'^^^W'

20

yj'j^ ^ -^ '^^ ''^•'' •'^ - ^ - ^ -^-^ -^ ^^J^.r :^ .^J'j^i^ j^ .-fy^j^jfj^j^j^j^ yy./^.yjfj .J^.^ff

10

. . j V ^ y Average Percentage of Resources invested In Hardware Davelopmerrt

C/Z

1991 19S2 1393 1994 1995 1996

1997

Source: Dataquest (November 1991)

Figure 2

Current Distribution of Investment of Resources for Hardware and Software Development

Percentage of Electronic Design Market

50-

40

30

20.

10-

26.7

25.7

23,3 22,9

20.5

20.8

15.3

4.6

10.6.

1.2

13.2

9.8

10.7

12.6

6.5

7.3

5.2

0 1-10 11-20 21-30 31-40 41-50 51-60 61-70 71-80 81-90 91-100

Percentage of Resources Invested in Hardware and Software Development

Percentage of Resources Invested | I Percentage of Resources Invested

In Software Development In Hardware Development

Source: Dataquest (November 1991)

©1991 Dataquest Incorporated / 1290 Eidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

C A D / C A M / C A E — E l e c t r o n i c Design A u t o m a t i o n

Figure 3

Design Productivity Gap

Chip Complexity

100X|-

90X

SOX

1 1

H I IC Integration Capability

^ ^ H Design Productivity

70X

60X

SOX

40X

SOX

20X

23X

92X

1

Design Productivity Gap =

ESDA Opportunity*

10X

IX

1X

4.3X^

5X

~3

1980

?X

1985 1990

1995

*To fill gap, engineers will turn to better design automation tools and Increased software content to exploit the vast quantity of high-performance VLSI and ULSI hardware.

Source: Dataquest (November 1991)

Figure 4

The Evolution o f EDA

Level of Design Automation

High

Automation of

System Design

Automation of

Logic Design

Automation of

Physical Design

L o w

1970

Source: Dataquest (November 1991)

1980

1990 2000

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-6000 / Fax (408) 437-0292

0011644

CAD/CAM/CAE—Electronic Design A u t o m a t i o n

Figure 5

CASE/Software Tools Market Evolution

Market Penetration Percent

7 . 0 -

Point Products Limitations

• Laci< Integration

• Incompiete Functionality

• Ineffective interfaces

• Difficult to Maintain

Integrated Case

• Development Environments and Frameworks

• Object-Oriented Technology

• Distributed and interoperable Systems

S3.4

Btiiion

Market

$4.2

Billion

Market

7%

$5.2

Billion

Market

(Actual)

Period of

Consolidation

5 . 0 -

$13.0

Billion

Market

/

(Estimated)

Missionary

Selling

1975

Source: Dataquest (November 1991)

1988 1989 1990 the market with CAE tools for schematic capture, logic synthesis, and design verification.

Meanwhile, CASE companies such as Czdie

Technologies Inc., i-Logix Inc., and Knowledgeware Inc. began providing automated tools for the analysis, design, and simulation of the software systems.

In sum, the EDA industry has evolved from the automation of the physical design in the 1970s to the automation of logic design in the 1980s.

Similarly, the CASE industry has evolved from the automation of code generation (via advanced compilers and higher-level languages) in the 1960s and 1970s to the automation of design, analysis, and requirements capture in the 1980s. What will the 1990s hold? Dataquest believes that it will be the era in which the market demands tighter integration between

EDA and CASE to automate the electronic system design process.

Dataquest Perspective

Dataquest believes that the military and aerospace industries, as well as most major computer manufacturers, will be at the forefront in

1992

1995 adopting ESDA. These industries have been early adopters of both EDA and G\SE tools and generally demonstrate the sophistication level required to implement ESDA technology.

Although a range of companies in those industries have expressed a strong interest in ESDA tools, a primary consideration is just how quickly they can absorb this emerging species of design automation. Among the barriers to

ESDA's adoption are a lack of perceived need, lack of standards, software and hardware communications barriers, incompatible development environments, end-user sophistication level, and the inertia accompanying today's electronic design process. Inadequate internal organizational structures initially will also inhibit market growth.

Currendy, most electronics manufacturers do not fully comprehend ESDA, primarily because

ESDA remains in its embryonic stage. Thus, the challenge for ESDA tool suppliers will be to demonstrate the high productivity gains that can be achieved with the technology. Reducing both time to market and the cost of products developed while increasing product functionality are

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

CAD/CAM/CAE—Electronic Design Automation the most important goals of the market, according to Dataquest's most recent research.

Dataquest believes that the migration toward

ESDA will follow an evolutionary rather than a revolutionary path. Our research currently shows that there are very few sites that have adequately automated the electronic systems design process. In fact, our research has been unable to uncover a site that is fully satisfied with its electronic design process. market's aversion to designing complex, mixed-signal circuits. Yet there are strong signs that the mixed-signal EDA market opportunity will start to tmfold in earnest over A e next 12 to 18 months and begin bearing the fruit that it has long promised. Our view is based on the convergence of a range of enabling technologies and market forces applicable to mixedsignal design. Figure 1 illustrates the projected growth of the mixed-signal design verification market.

Is the mainstream systems design market ready for this technology now? At this time, the answer clearly is "no," but the ESDA market should begin to blossom by mid-1993. Creating and developing a market for ESDA products is crucial to vendors succeeding with these tools in the future. It is important to note that maintaining compatibility with the current generation of design automation tools will be a key ingredient for success in the impending ESDA market. •

By Ron Collett

John DeArmon

Historical Obstacles to Growth

EDA and ASIC vendors alike have made attempts to penetrate the mixed-signal design market in the past 5 to 10 years. The mixedsignal computer-aided engineering (CAE) market was less than $20 million in 1990 and has been growing at only 15 to 20 percent compound annual growth rate (CAGR) during the past five years. The mixed-signal ASIC market, by contrast, grew at a CAGR of 25 to 30 percent from

1985 to 1990 and reached $400 million in 1990.

Despite this ostensibly robust growth, the mixed-signal ASIC market has also fallen short of expectations. This is because the design-start growth rates have been in the 14 to l6 percent range. Thus, the mixed-signal ASIC market can be charaaerized as having a small number of designs that are produced in high volume.

Interpreting the Mixed

Signals of the Mixed-

Signal Market

Capitalizing on the mixed-signal market opportunity has been a tough challenge for both

ASIC and tools suppliers alike. (In this context, mixed-signal refers to electronic systems comprising a mixture of digital and nondigital, or analog, circuits.) It is a market that continues to elude both camps. For example, high cost coupled with high risks and long development cycles, not to mention severe test problems, have caused much of the potential market either to forego developing products requiring advanced mixed-signal ASICs, develop less ambitious designs, and/or use an alternative solution such as discrete analog and digital ICs mounted to a printed circuit board (PCB). As a result, the mixed-signal design market has remained smaller than anticipated, and the vast majority of mixed-signal PCB and ASIC designs have remained fairly simple.

The mixed-signal electronic design automation

(EDA) market has been a casualty of the

Significant reasons why potential customers have shied away firom mixed-signal ASICs include high cost, high risk, weak design tools, and device-testing difficulties.

CMOS ASIC vendors have attacked the mixedsignal market by offering macrocell libraries that typically include simple, low-performance analog functions that can be combined with digital functions onto a single semicustom IC. But with these chips, the designer has traditionally had to sacrifice performance. The chips offer very good digital performance but only fair analog capabilities. They do not combine the performance and functionality advantages of dissimilar

IC fabrication processes (for example, bipolar and MOS) onto a single semicustom device.

However, with the recent advances in CMOS and BiCMOS fabrication processes, designers

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

8

CAD/CAM/CAE—Electronic Design Automation

Figure 1

The Mixed-Signal Opportunity—Projected Growth of the Mised-Signal Design

Verification Market

i

Source: Dataquest (November 1991) are finding that fewer performance sacrifices are necessary.

Yet BiCMOS fabrication processes have only recently come to the forefront and are still costly. In addition, the volumes for mixedsignal BiCMOS chips are likely to remain modest given the alternative technologies that are emerging, namely, advanced CMOS processes and thin film multichip modules.

Meanwhile, bipolar ASIC vendors have attacked the market with tile array-based solutions.

Although this technology offers maximum analog performance and capabilities, the amount of digital logic that can be integrated onto the device is minimal. In addition, because bipolar transistors are used to implement the digital logic, there is a power consumption penalty.

Beyond unacceptable design trade-offs, other significant reasons why potential customers have shied away from mixed-signal ASICs include high cost, high risk, weak design tools, and device-testing difficulties. The end result for mixed-signal ASIC suppliers has been lower gross volumes, a small market, and potential markets that remain undeveloped.

Not surprisingly, the response from EDA vendors has been to channel tool development resources elsewhere. Consequently, the mixedsignal EDA tools available thus far have not fully delivered what the market wants: fast, easy-to-use, highly accurate verification tools that tightly integrate the analog and digital design domains. Similarly, lack of a robust, easy-to-use standard analog modeling language also has impeded growth.

fust as the 1980s can be thought of as the era of personal computing, our belief is that the 1990s will be characterized as the decade of embedded control, telecommunications, and data communications.

The application-specific standard product (ASSP) mixed-signal market, by contrast, has been stronger than the mixed-signal ASIC market.

This stems from the fact that ASSP makers typically have not only greater design capability and design methodology expertise, but also because an ASSP's development costs and risks can usually be amortized across potentially higher volumes. We expect the mixed-signal

ASSP market to enjoy even stronger growth in the next five years.

©1991 Dataquest Incorporated / 1290 Kdder Park Drive, San Jose, CA 95131-2398 / (408) 437-6000 / Fax (408) 437-0292

0011644

i

CAP/CAM/CAE—Electronic Design Automation

Because the mixed-signal EDA market opportunity has remained fairly small, EDA vendors have been wise to channel resources elsewhere.

However, the landscape of the mixed-signal design market is changing. We anticipate that the mixed-signal design market will soon move into a period of significant expansion. As such, the time is ripe for EDA vendors to begin boosting their mixed-signal product development efforts.

Mixed-Signal Application Market

Drivers

Just as the 1980s can be thought of as the era of personal computing, our belief is that the

1990s will be characterized as the decade of embedded control, telecommunications, and data comutnunications. Voice and data communications markets will develop more fully throughout the decade and will provide new opportunities for semiconductor suppliers and EDA tool vendors looking to serve the mixed-signal design market. Embedded control requiring mixed-signal circuits will also suige to the forefront in automotive, industrial control, and consumer electronics applications.

Specific examples where mixed-signal technology will find increasing use include the following:

• Digital cellular telephones

• Digital video

O HDTV

O Facsimile

O Display digital-to-analog converters (DAC^)

• Medical-imaging systems

• Automotive control

• Mass storage devices

• Digital audio

O Compact discs

D Digital audio tape

• Multimedia

• Modems

• Automatic test equipment

Technology Enablers

Application markets represent only one driver in the mixed-signal market paradigm. Several technology enablers are also at the heart of the mixed-signal design market.

For example, a vehicle that is more effective than either ASICs or traditional PCBs for carrying mixed-signal circuitry is needed to give the mixed-signal design market a solid boost. In other words, a technology is needed that offers the low risk, fast turnaround time, and low cost of PCB-based mixed-signal design, yet also delivers performance and integration that begins to approach that of an ASIC. In short, it must be a technology that appeals to the mainstream.

We believe that multichip modules (MCZMs) will offer the market this exact combination, albeit not immediately.

With MCMs, the mixed-signal market receives the best of all worlds: high integration, high performance, reduced risk, lower costs, and less complex testing requirements.

We believe that MCMs will become a primary workhorse for mixed-signal systems in the second half of the decade. As an alternative to combining analog and digital circuitry on a single ASIC to achieve high integration and high performance, designers can package analog and digital bare dice in a single MCM. This allows the user to optimally mix dissimilar semiconductor febrication technologies. Thus, with MCMs, the mixed-signal market receives the best of all worids: high integration, high performance, reduced risk, lower costs, and less complex testing requirements (provided that bare die can be adequately tested prior to being attached to the MCM). For many applications the result will be integration and performance advantages that, although not equivalent, approach the advantages of an ASIC. Moreover, the risks and costs associated •with this approach will be dramatically lower than those of an ASIC implementation.

Arguably, perhaps, traditional hybrid devices could have provided the basic vehicle to fuel the growth of the mixed-signal design market in the past. After all, a hybrid certainly enables the designer to mix bare chips manufactured with dissimilar semiconductor fabrication

©1991 Dataquest Incoipoiated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

10

CAD/CAM/CAE—Electronic Design Automation technologies. Why didn't this happen? Simply stated, hybrids have not and do not afford the high interconnect density of an MCM. And high interconnect density is a key requirement for the laigest segment of the mixed-signal design market—those designs having a large digital content and a small analog content. Dataquest's research indicates that an average design's functionality can be attributed to 82 percent digital capability and 18 percent analog capability (see

Figure 2). Thus, although hybrids allow for the mixing of dissimilar IC technologies, they do not match the high integration needs of much of the market. MCMs, by contrast, do meet the market's needs in both areas.

The MCM-based mixed-signal market will not blossom until MCMs do in fact move doum the price learning curve.

High interconnect density, but more importantly, improved system performance, are primary drivers fueling MCM demand. Initially, MC^Ms will be driven down the price learning curve primarily by computer manufacturers' demand for higher speed-enabling technologies, which is an important characteristic of MCMs. The high integration that MCMs afford, initially, is a distant secondary advantage in most instances.

Integration advantages of MCMs will increase market pull toward the middle of the decade.

Moreover, once the price of an MCM reaches approximate parity with the traditional PCBbased solutions (likely in the second half of the decade), the path will have been completely cleared for mixed-signal MCMs to move into the mainstream. It is important to note, however, that the MCM-based mixed-signal market will not blossom until MCMs do in fact move down the price learning curve. We expect expansion of the MCM-based mixed-signal design market to be a valuable by-product of the strides made in MCM technology. Yet mixed-signal design demands will not be the primary driver behind

MCM. growth in the short term.

Other technology drivers that portend to fuel mixed-signal design growth include the following:

Improvements in CMOS processes, which enable analog functions to be implemented using field effect transistors

BiCMOS fabrication process advancements

Improvements in CAE technology (for example, simulation and modeling)

Figure 2

Projected Allocation of System Functionality (Digital versus Analog)

Percentage Allocation

100

1991

19S2

Source: Dataquest (November 1991)

1993

1994

1995

1996 1997

©1991 Dataquest Incorporated / 1290 Kidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

CAD/CAM/CAE—Electronic Design Automation 11

The availability of automatic layout tools for analog design will also help fuel growth. These tools will soon begin migrating from the research laboratory to the commercial market.

Competitive Market Forces

Advancements in manufacturing technologies and the growth of application markets are just two of several forces that will converge and drive the mixed-signal EDA market. Others include increasing demand from the mixed-signal ASSP and ASIC markets, both of which will experience very strong growth during the next five years. Mixed-signal ASSP market growth will be driven by semiconductor manufacturers wanting to put analog circuitry on chip as a way to differentiate their product offerings. Integrating analog circuitry that traditionally has existed separately from the standard IC (for example, converters, amplifiers, and drivers) differentiates product offerings. Digital ASSP manufacturers vwll also be competing against digital ASIC manufacturers that- are enhancing their megacell libraries to include functions that traditionally have been sold as ASSPs and off-the-shelf ICs.

Including analog on chip helps this camp compete more effectively against the digital ASIC players.

To date, no EDA company has captured a dominant position in terms of market perception in the mixed-signal EDA market.

ASSP manufacturers will also benefit from the steady improvements being made in CMOS and

BiCMOS fabrication processes. Lower costs and improved performance associated with these technologies will make mixed-signal ASSPs more attractive to semiconductor suppliers that hope to expand traditional markets as well as move into new markets.

Another force that will help expand the mixedsignal EDA market is the trend toward using digital logic to implement functions that traditionally have been designed with analog circuits. Fewer analog circuits within a design eases the burden on mixed-signal verification tools. This, in turn, should enable mixed-signal verification tools to more easily tackle the upcoming generation of mixed-signal designs.

As the mixed-signal design market evolves during the next five years, we believe that the following market segmentation provides a useful model:

• PCB-based mixed-signal designs, which combine purely discrete analog and digital ICs

• Mixed-signal IC designs that are customized to the designer's exact requirements and are customer-specific products (These designs are typically implemented through the use of a macrocell library and are commonly referred to as mixed-signal ASICs.)

• Multichip module- and hybrid-based mixedsignal designs, which combine purely analog and digital ICs, and perhaps some mixedsignal ASICIs, all of which are unpackaged

• Mixed-signal IC designs in the form of ASSPs

(these products integrate the analog and digital functionality into a single IC or hybrid, which is sold as a standard product.)

Each of these four categories can be further subdivided into three distinct groups:

• Mixed-signal designs that are mosdy digital

• Mixed-signal designs that are mosdy analog

• Mixed-signal designs that have nearly equal amounts of analog and digital circuitry

(Finer segmentation than this is certainly feasible, but for the moment, this market model is sufficient.)

Figure 3 shows the projected size of each segment of the mixed-signal EDA market in the

1996 time frame.

Dataquest Perspective

Figure 4 illustrates the combination of forces that will converge and drive the mixed-signal

EDA market. To date, no EDA company has captured a dominant position in terms of market perception in the mixed-signal EDA market.

This contrasts sharply with other segments of the EDA market, where particular EDA suppliers have carved out strong positions. Thus, the time

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

12 CAD/CAM/CAE—Electronic Design Automation

Figure 3

Projected Djstributioii of the Mixed-Signal EDA Market (1996)

^ ^ I m p l e m e n t a t i o n

D i g i t a l - " ^ ^ ^

Analog ^ v ^ ^

PCB

Mix ^ \ ^

Mostly

Digital

16

ASSP

29

Percentage

ASIC

23

MCM

9

Mostly

Analog

Approximately

Equal

Analog-

Digital

4

1

Total 21

_

Source: Dataquest (November 1991)

5

2

36

3

3

29

1

4

14

Total

77

13

10

100

Figure 4

Forces Acting on the Mixed-Signal Design Market

Emerging Application Markets

Advances In CMOS Manufacturing Advances in BiCMOS Manufacturing

ASSP Differentiation

ASSP vs. Megacell Differentiation

Advances in Analog Automatic

Layout Tools

Improved Mixed-Signal CAE Tools

Availability of MCMs

Source: Dataquest (November 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

CAD/CAM/CAE—Electronic Design Automation

13 is ripe not only to invest resources in technology development but also in market positioning. •

By Ron Collett

The New Business

Model of the 1990s for EDA Vendors

In order to compete in the hody contested electronic design market of the future, latge electronic design automation (EDA) software vendors must begin making fundamental changes in their current business models.

Whereas in the past laige EDA vendors developed or acquired (by corporate acquisition) nearly all products that they brought to the mariset, Dataquest believes that this strategy will be ineffective for companies hoping to achieve annual software and service revenue of more than $350 million to $4(X) million.

Dataquest believes that vendors reaching annual sales in this range must shift their focus from

Figure 1

Future Structure of the EDA Industry

Third-Party Tool Developers

"wall-to-wall" product development and ownership to wall-to-wall product distribution. In other words, large EDA vendors must provide not only internally developed tools but also aggressively distribute a broad range of products from third parties.

Our conclusion stems from the belief that large

EDA vendors will not be capable of sustaining a competitive edge across their complete portfolios of products and technologies. In the face of emerging framework standards, it will become less difficult for third parties to integrate their products into the customer's design environment. Hence, the competitive landscape of the EDA industry is changing (see Figure 1).

The New Business Model

Briefly, the new business model that Dataquest proposes is for EDA vendors to gracefully forego ownership of all nonstrategic product lines. In this context, nonstrategic refers to products that neither offer high margins nor engender substantially better account control.

Conversely, products that exhibit either of these characteristics should be owned. It goes without saying that a prerequisite charaaeristic for a iieDA

:;:VAB:

VAR-Developed Design Automation Tools

Third-Party Design Automation Tools

Framework Technology

Integration Services

Consulting Services

Customer Service and Support

i

"Worldwide

|:-Sectroni&

Ip'^besign

f Market

1-

•• iii'

Source: Dataquest (November 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

14 CAD/CAM/CAE—Electronic Design Automation product that an EDA supplier wishes to own is that it must offer the vendor a substantial and sustainable competitive advantage in the market.

No Standard set of industrywide specifications exist under which

EDA vendors can construct standard environments for EDA software tools, although the situation is changing rapidly.

to remain focused and attract top-flight developers; the combination translated into an ability to develop competitive products. And, finally, service and support from a single entity was an attractive lure.

For those technologies that EDA vendors could not develop internally, many of these companies turned to the outright acquisition of companies with desirable technology. Hie ensuing spate of acquisitions resulted in much laiger companies that have found it increasingly difficult to quickly deliver innovative software products to the market.

Although EDA products traditionally have moved through the direct sales channel with very little OEM distribution, Dataquest believes that large EDA vendors must look toward leveraging the power of their distribution channels and begin distributing the tools of those thirdparty vendors with products offering better capabilities. In the past, most distribution relationships, which fiave come in the form of

OEM deals, have failed because they were stopgap measures aimed at temporarily filling a product void.

The most successful large EDA vendors in the future (1995 and beyond) will be able to seek out advanced design automation technology from both niche suppliers and large electronics manufacturers that want to commercialize their internal tools. The EDA vendor's responsibilities include productizing and integrating the technology. What this means for many EDA vendors is a basic change, or restructuring, of the way they do business.

Evolution o f t h e P r o b l e m

Between 1985 and 1990, the business model of choice among rapidly growing EDA vendors was to be a full-line supplier. In this context, full-line supplier implied complete ownership of the EDA technology. Dataquest believes that the full-line supplier was a correct one for that time period for several reasons. First, very few standards were in place for mixing and matching tools from different suppliers. Thus, reasonable integration among tools could only be achieved by purchasing tools from a single source. Second, locking out small niche suppliers enabled large EDA vendors to capture larger market share and thus maximize revenue. Third, even the laigest EDA vendors were still small enough

The main catalyst behind the recent flurry of OEM partnerships is that it is becoming far too complex and difficult for EDA companies alone to provide all of the necessary design automation tools to the market.

while larger EDA companies have had problems developing products, many smaller companies (and laige electronics manufacturers hoping to merchandise their in-house tools) simply lacked the necessary marketing and sales muscle required to compete successfully against more established competitors. Both camps will soon need each other to achieve market success. In the coming years, tool vendors must be able to deliver advanced EDA software even more quickly to keep pace with advances in chip manufacturing.

Leveling t h e Playing Field

To date, no standard set of industrywide specifications exist under which EDA vendors can construct standard environments for EDA software tools, although the situation is changing rapidly. For instance, hardware description languages, such as Verilog and VHDL, are becoming increasingly common vehicles for model development. Emerging framework standards such as those fi-om the CAD Framework

Initiative (CFI) as well as de facto framework standards will also become a crucial element in the integration of EDA software in the next few years. As standards emeige, niche players will be able to develop software conforming to

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

CAD/CAM/CAE—Electronic Design Automation

15 these standards and, hence, will be able to compete more aggressively against the laiger suppliers. In essence, standards will begin to level the EDA playing field.

Changing the Business Model

What should EDA vendors' new business model look like? In order to ensure competitiveness throughout the impending decade, we believe that large EDA vendors' business models should include the following key elements:

• Owning and/or developing only strategic core technologies (for instance, those that significandy improve account control and/or yield sustainable high profit margins)

• Aggressively integrating and distributing thirdparty software tools

• Providing a framework that integrates the various disparate design automation products

• Providing the support for third-party tools, thus maintaining the primary interface with the customer to better ensure account control

• Establishing technology partnerships with third-party design automation suppliers

In addition, EDA vendors should strive to cultivate the resources in-house that will seek out new design automation tools to move through their channel. Some EDA companies have begun taking this approach. For example,

Dazix/Inteigraph recendy signed an exclusive six-year technology licensing agreement with

AT&T Microelectronics of Berkeley Heights, New

Jersey, under which Dazix will incorporate

AT&T's internal design synthesis technology into its future EDA products. Another example is an agreement between Mentor Graphics Corporation and Texas Instruments Inc. (H) under which the companies exchanged technology on design-for-manufacturability software. Mentor agreed to sell TI's design-for-manufacture (DFM) products, and TI agreed to integrate the tools into Mentor's Falcon Framework. Likewise,

AT&T and Cadence Design Systems Inc. collaborated on a similar DFM deal.

Other recent examples of companies striking similar deals include OEM deals between

C:adence and AT&T; Quad Design Technology and Mentor Graphics; and agreements by Teradyne Incorporated with Racal-Redac Inc. and

Vantage Aiialysis Systems Inc. Such third-party technology will come into play more often in the future, with successful EDA vendors productizing technology developed by large electronics manufacturers, niche start-ups, and perhaps even research institutions.

Dataquest Perspective

The main catalyst behind the recent flurry of

OEM partnerships is that it is becoming far too complex and difficult for EDA companies alone to provide all of the necessary design automation tools to the market. During the past five to six years, EDA companies typically have addressed this need through outright corporate acquisitions. But, it is becoming dear that product ownership does not necessarily equate to market success. Thus, it is now strategically important for large EDA companies to begin distributing and joindy marketing more tools from outside sources and third-party developers.

In sum, to provide their customers with the most value, large EDA vendors need to become solution suppliers and, in a sense, value-added resellers. The most effective suppliers will be able to take the best technology from a wide variety of outside sources and integrate such technology into effective EDA solutions. •

By Ron Collett

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

16

In Future Issues

Watch for reports on the following topics in future issues of Dataquest Perspective:

• Update of European EDA market

• Overview of IC layout market

CAD/CAM/CAE—Electronic Design Automation

I

I

For More Information . • .

Managing Editor Patrick Waurzyniak (408) 437-8132

On the topics in this issue CAD/CAM/CAE (408) 437-8132

On related subjects Client Inquiry Center (408) 437-8671

About other Dataquest publications Sales (408) 437-8000

About upcoming Dataquest conferences Conferences (408) 437-8245

About your subscription Customer Service (408) 437-8402

Via fax request Fax (408) 437-0292

Tlie contesit of diis report represencs our interpretation and analysis of information generally available to the public or released by responsible individuals in the subject con^>anies, but is not guaranteed as to accuracy or completeness. It does not contain material provided to us in confidence by our dients. Individual companies reported on and analyzed by

Dataquest may be clients of this and/cM* other Dataquest services. This information is not fiuni^ied in connecdon with a sale or o^er to sell securities or in connection with the solicitation of an offer to buy securities. This firm and its parent and/or their officers, stockholders, ormonhersoftheirfiuniliesmay, from time K> time, have a long or shon position in the securities mentioned and may s ^ or buy such securities.

©1991 Dataquest Incoipoiated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011644

i

Dataquest

n n acompanyoF

MMMM ThcDun&BnidstnxlCwpDration

Dataquest

Perspective

CAD/CAM/CAE

Index

June-September 1991 October 25, 1991

How to Use This Index

This is a cumulative index of key terms, personal names, and company names for the first two quarters of

Volume 1 of Dataquest Perfective. Entries are followed by the date of publication and the F>age number(s).

Boldface type indicates major topic titles. Produa names are listed under the company that manufactures/publishes the product General information about the companies themselves is found under the full company name.

Each citation indicates only the beginning page of a discussion of a topic (the range of p>age numboB is not cited).

Dataquest Perspective issues covered i n this index:

VoL 1, No. 1: June 24, 1991

A View from 30,000 Feet: CAD/CAM/CAE versus the World Computer Market, 2

Schlumberger Jumps on Hewlett-Packard's Workstation Ride, 4

Object-Oriented Databases: A Technology Needing a Kick Start, 8

EDA Opp>ortunities in the New Design

Paradigm, 12

Computervision Offers Major Upgrade to CADDS

Product Line, 14

VHDL International Consortium Formed by EDA

Vendors, 15

VoL 1, No. 2: July 8, 1991

A View from 15,000 Feet: Looking at the Landscape of the Entire CAD/CAM/CAE Industry, 2

Computervision's CADDS 5: More than Just a

Face-Lift, 4

What Windows Means in the Long Term, 7

Hewlett-Packard to Port Mechanical-Design Software to Sun SPARCstation, 10

ESRI Goes GUI, 10

Intergraph's Dazix Subsidiary Signs Technology

P a a with AT&T, 11

VoL 1, No. 3: July 22, 1991

Apple and IBM Join Hands: Do Two Proprietaries

Equal an Open? 2

MS-DOS 5.0: Leaner, Faster, Stronger, Better, 4

Digital Equipment Reenters the PC Market with a

Bang, 8

VoL 1, No. 4 : August 26, 1991

Impending Framework Fragmentation Triggers New

EDA Business Strategies, 2

Silicon Graphics Announces Indigo, a New Low-

End Workstation for Less than $10,000, 4

Highland Software Ships Software Store, 7

Workstation Vendor Strategies in the DIMS

Market—Sun Microsystems, 9

Cadence and Siemens Nixdorf Enter Joint EDA

Venture, 14

©1991 Dataquest Incorporated / 1290 Kidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011617

CAD/CAM/CAE

Advanced Computing Environment (ACE)

Digital DECpc 433 Workstation and, (Jul 22):10 status of, Qui 8):8

AEC. See Architecture, engineering, and construction

Apple adoption of IBM RS/6000 POWER architecture,

Oul 22):2 company. See Apple Computer Inc.

Macintosh integrated into IBM's computer systems,

Oul 22):2

QuickTime multimedia environment, 0ul 22):2

Apple Computer Inc.

CAD/(>M/CAE revenue of, 0ul 8):3

IBM's relationship with, 0ul 22):2 market share, CAD/CAM/CAE, Oul 8):3

Architecture, engineering, and construction

(AEC)

worldwide CAD/CAM/CAE market for, Oul 8):2

Asia/Padfic-Rest of World (ROW)

CAD/CAM/CAE market in, Oul 8):2

EDA framework market in, (Aug 26):3

ASICs design reuse for, Oun 24):13 gate arrays, cost per gate, Oun 24):12 gate densities for, Oun 24):12 hardware megacells for, Oun 24): 14

HDL design of, Oun 24):13

AT&T

Design Verification System, Oul 8): 11

GENTEST test-vector generation tool, Oul 8):12

High-Level Design System, Oul 8):11

Test-Scan System, Oul 8):11

Testpilot software, Oul 8): 11

AT&T (American Telephone and Telegraph Co.)

Dazix agreement with, Oul 8): 11

Autodesk Inc.

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, Oul 8):3

Cadence Design Systems Inc. (continued)

EDA framework market of, (Aug 2^:3 market share, CAD/CAM/CAE, Oul 8):3

Open Verilog International (OVI) consortium,

Oun 24):15 ports to SGI IRIS Indigo workstation, (Aug 26):7

Siemens EDA joint venture with, (Aug 26):l4

Cadence Europ>e GmbH formation of, (Aug 26): 14

Calera Recognition Systems

Sun's imaging software partnership with,

(Aug 26):12

CD-ROM distribution, (Aug 26):7

CFI (CAD Framework Initiative), (Aug 26):2

Company analysis

Digital, Oul 22);8

Sun Microsystems, (Aug 26):9

Compaq Computer Corp.

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, Oul 8):3 worldwide desktop unit shipments, Oul 22):9

Computer industry

worldwide trends for, Oun 24):2

Computervision

CADDS 4X, Sun SPARCstation upgrade for,

Oul 8):4

CADDS 5

Design & Drafting 3-D design tool, Oul 8):5

DesignView variational sketcher, Oul 8):4 overview of, Oun 24):14

Parametric Design module, Oul 8):5

Premium Engineering Package, Oul 8):4 product analysis of, Oul 8):4

Solid Modeling design tool, Oul 8):5

View & Markup module, Oul 8):5 company. See Computervision (comjsany)

Computervision (company)

CAD/CAM/CAE revenue of, Oul 8):3

CAD/CAM/CAE revenue tracking improved,

Oul 8):4 market share, CAD/CAM/CAE, Oul 8):3 software revenue on technical workstations,

Oun 24):7

Construction. See Architecture, engineering, and construction

Control Data Corp.

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, Oul 8):3

CAD/CAM/CAE industry

Dataquest anniial delivery schedule for, Oul 8):3

Dataquest market share database, Oul 8):4 market analysis of, Oun 24):2; Oul 8):2 unbundling of MCAE/CAD/CAM software,

Oul 8): 10 worldwide market portfolio for, Oul 8):2 worldwide trends for, Oun 24):2

CAD Framework Initiative (CFI), (Aug 26):2

Cadence company. See Cadence Design Systems Inc.

Design Framework H, (Aug 2®: 14

Verilog Hardware Description Language (VHDL),

Oun 24): 15

Cadence Design Systems Inc.

CAD/CAM/CAE revenue of, Oul 8):3

D

Dangermond, Jack, Oul 8): 11

Dataquest annual delivery schedule, Oul 8):3

Dataquest market share database, Oul 8):4

Dazix (company)

AT&T agreement with, Oul 8):11

Dazix/Intergiaph

EDA framework market of, (Aug 26):3

Demonstrations

software, cost of, (Aug 26):8

©1991 Dataquest Incorporated / 1290 Eidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011617

i i i

CAD/CAM/CAE

Desktop computers

worldwide shipments of, (Jul 22):9

Digital company. See Digital Equipment GDrp.

DECpc 320SX notebook, (Jul 22):9

DECpc 333 portable, Qui 22):9

DECpc 433 Workstation, Qui 22): 10

DECpc 433T deskside computer, (Jul 22):10

DECpc "D" series, (Jul 22):9

DECp>c line series overview, (Jul 22):9

DECpc "N" series, (Jul 22):9

DECpc "P» series, (Jul 22):9

DECpc "T" series, Qui 22):9

DEC^station 50(X), performance of, (Jun 24):5

Digital Equipment Corp.

CAD/CAM/CAE revenue of, (Jul 8):3 company analysis of, (Jul 22):8

EDA framework market of, (Aug 26) :4 market share, CAD/CAM/CAE, (Jul 8):3 strategy of, (Jul 22):8 worldwide desktop unit shipments, (Jul 22):9

DIMS. See Document image management systems

(DIMS)

DIP (Document i m ^ e processing). See Document image management systems (DIMS)

Distribution methods

CD-ROM, (Aug 26):7

Document i m ^ e management systems (DIMS)

industry division for, (Aug 26) :9

Sun Microsystems' market strategy (company analysis), (Aug 26):9

Fujitsu Ltd.

CAD/CAM/CAE revenue of, (Jul 8):3 market share, CAD/CAM/CAE, (Jul 8):3

Gate arrays. See under ASICs

Gates, Bill, Oul 8):9

Geographic information systems (GIS)/mappii^

ESRI ARC/INFO 6.0, Oul 8): 10 worldwide CAD/CAM/CAE market for, Oul 8):2

GIS. See Geographic information systems (GIS)/ mapping

GUIs

impact on GIS product requirements, Oul 8): 10

ECAD. See Electronic design automation (EDA)

ECAE. See Electronic CAE

EDA. See Electrotuc design automation (EDA)

Electronic CAE

worldwide CAD/CAM/CAE market for, (Jul 8):2

Electronic d e s ^ n automation (EDA)

Cadence-Siemens Nixdorf joint venture for European market, (Aug 26): 14 framework fragmentation of, (Aug 26):2

HDL-based design reuse for, (Jun 24):12

Intergraph's Dazix subsidiary agreement with

AT&T, Oul 8):11 market analysis of, (Aug 26):2 productivity analysis of, (Jun 24): 12

Siemens Nixdorf-Cadence joint venture for European market, (Aug 26): 14 technology analysis of, Qua 24):12

VHDL as standard for, 0un 24):15

Engineering. See Architecture, engineering, and construction

ESRI

ARC/INFO 6.0, 0ul 8);10

ArcView geographic query-and-answer software,

(Jul 8): 11

Europe

CAD/CAM/CAE market in, Oul 8):2

EDA framework market in, (Aug 26):3

H

Hardware description language (HDL)

ASIC design reuse with, Oun 24):13

Verilog HDL, Oun 24): 15

VHSIC (VHDL), Oun 24):15

HDL. See Hardware description language (HDL)

Hewlett-Packard Co.

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, Oul 8):3 products of. See HP

Schlumbetger agreement for Apollo 90(X) Series

700 and Series 400 workstations with, Oun 24):4 software revenue on technical workstations,

Oun 24):7

Highland Software

Software Store, product analysis of, (Aug 26):7

Hitachi Ltd.

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, Oul 8):3

Host-based terminals

worldwide CAD/CAM/CAE market for, Oul 8):2

Host-dependent computers

trends for, Oun 24):2

Host-dependent terminals

HP worldwide CAD/CAM/CAE market for, Oul 8):2

Apollo 9(X)0 Series 400 workstations, produa analysis of, Oun 24):4

Apollo 9000 Series 700 workstations

Model 720, Oun 24):5

Model 730, Oun 24):5

Model 750, Oun 24):5 product analysis of, Oun 24):4 company. See Hewlett-Packard Co.

Data Management System, Oul 8):10

DNIOK PRISM-based workstations, Oun 24):6

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011617

CAD/CAM/CAE

HP (continued)

IGES translator, Oul 8):10

ME 10 2-D design and drafting software, (Jul 8):10

ME 30 solid modeling software. Qui 8): 10

Precision Architecture (PA)-RISC architecture,

Oun 24):4

Hybrids. See Printed circuit boards/hybrids/multichip modules (PCBs/hybrids/MCMs)

IBM

ADC operating system, Qui 22): 2 company. See IBM Ck>rp.

RS/6000 Model 320H, performance of, 0un 24):4

RS/6000 POWER architecture, (Jul 22):2

IBM Corp.

Apple's relationship with. Qui 22):2

CAD/CAM/CAE revenue of. Qui 8):3 market share, CAD/CAM/CAE, (Jul 8):3

MiCTOSoffs relationship with, (Jul 8):8; Oul 22):2

Motorola's relationship with, (Jul 22):3 worldwide desktop unit shipments, (Jul 22):9

ICs

application specific. See ASICs worldwide CAD/CAM/CAE market for, Oul 8):2

IEEE standards

1076 CVHDL), Oun 24):15

Intergraph Corp.

CAD/CAM/CAE revenue of, Oul 8):3

Dazix agreement with AT&T, Oul 8):11 hardware revenue tracking improved, Oul 8):4 market share, C::AD/CAM/CAE, Oul 8):3

Joint ventures

Apple-IBM software Of>eration, Oul 22):2

Cadence-Siemens Nixdorf EDA solutions for European market, (Aug 26): 14

IBM-Apple software operation, Oul 22):2

Siemens Nixdorf-Odence EDA solutions for European market, (Aug 26):14

JTS C^amputer Systems

Sun's imaging software parmership with,

(Aug 26):12

M

Mainframe computers

worldwide trends for, Oun 24):2

Mapping. See Geographic information systems (GIS)/ mapping

Market analysis

Apple-IBM joint venture software operation,

Oul 22):2

CAD/CAM/CAE industry, Oun 24):2; Oul 8):2

EDA framework fragmentation, (Aug 26):2

IBM-Apple joint venture software operation,

Oul 22):2

McDonnell Douglas Corp.

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, Oul 8):3 software revenue on technical workstations,

Oun 24):7

MCMs (multlchip modules). See Printed circuit boards/hybrids/multichip modules (PCBs/hybrids/

MCMs)

Mechanical applications

HP MCAD software ported to SPARCstation,

Oul 8):10 woridwide CAD/CAM/CAE market for, Oul 8):2

Mentor Graphics <^ip.

CAD/CAM/CAE revenue of, Oul 8):3

EDA framework market of, (Aug 26):3 market share, CAD/C:AM/CAE, Oul 8):3

Microsoft company. See Microsoft Corp.

MS-DOS and Microsoft's strategy for Windows, Oul 8):8 worldwide sales of, Oul 8):8

MS-DOS 5.0, product analysis of, Oul 22):4

OS/2 3.0 as a contribution to ACE, Oul 8):8 as a high-end system, Oul 8):7

Windows long-term strategy for, Oul 8):7 technology analysis of, Oul 8):7

Windows 3-0, sales of, Oul 8):7

Microsoft Corp.

IBM's relationship with, Oul 8):8; Oul 22):2

New Technology (NT) strategy of, Oul 8):7 operating system strategy of, Oul 8):7 strategy of, Oul 8):7

Midrange computers

worldwide trends for, Oun 24):2

Motorola Inc.

IBM's relationship with, Oul 22):3

Mouse devices

on portable computers, Oul 22): 10

Multichip modules (MCMs). See Printed circuit boards/hybrids/multichip modules (POs/hybrids/

MCMs)

Laptops, notebooks, and portables

80386-based portables, Qui 22):9

80386SX-based notebooks, Oul 22):9

©1991 Dataquest Incoipoiated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011617

i i i

CAD/CAM/CAE

N

NEC Corp.

CAD/CAM/CAE revenue of, (Jul 8):3 market share, CAD/CAM/CAE, 0ul 8):3

New Technology (NT) strategy, Qui 8):7

Nihon Unisys

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, (Jul 8):3

North America

CAD/CAM/CAE market in, Qui 8):2

EDA framework market in, (Aug 26):3

See also United States

Notebooks. See laptops, notebooks, and portables

NT (New Technology) strategy, (Jul 8):7

NYNEX Image Recognition Systems

Sun's imaging software partnership with,

(Aug 26):12

Product analysis

C:omputervision CADDS 5, (Jul 8):4

Highland Software's Software Store, (Aug 26):7

HP Apollo 90(X) Series 700 and Series 400 workstations, (Jun 24):4

MS-DOS 5.0, Oul 22):4

Schlumbetger-HP agreement for Apollo 9000 Series

700 and Series 400 workstations, Oun 24):4

Silicon Graphics IRIS Indigo workstation,

(Aug 26):4

R

Racal-Redac Inc.

EDA framework market of, (Aug 26)A

Rest of World (ROW). See Asia/Pacific-Rest of World

(ROW)

Resumix

Sun's imaging software partnership with,

(Aug 26):12

ROW (Rest of World). See Asia/Pacific-Rest of World

(ROW)

o

Object-oriented databases (OODBs)

technology analysis of. Qua 24):8

Object-oriented technology

Apple-IBM platform based on, (Jul 22):2

OCR Systems

Sun's imaging software partnership with,

(Aug 26):12

OODBs. See Object-oriented databases (OODBs)

Open VerUog International (OVI) consortium,

Oun 24):15

OVI (Open Verilog International consortium),

Oun 24):15

Padflc. See Asia/Pacific-Rest of World (ROW)

Page printers

Sun SPARCprinter, (Aug 26): 10

PCBs. See Printed circuit boards/hybrids/multichip modules (PCBs/hybrids/MCMs)

PCs. See Personal computers

Personal computers (PCs)

Sun SPARCstations vs., (Aug 26):11 trends for, Qua 24):2 worldwide C:AD/CAM/CAE market for, (Jul 8):2

Plexus Software Inc.

Sun's imaging software partnership with,

(Aug 26):12

PM. See Presentation Manager (PM)

Portables. See Laptops, notebooks, and portables

Presentation M a n n e r (PM)

Windows as an alternative to, (Jul 8):7

Printed circuit boards/fayforids/multichlp modules

(PCBs/hybrids/MCMs)

woridwide CAD/CAM/CAE market for, (Jul 8):2

Schlumberger company. See Schlumberger Technologies

DesktopBravo! CAD/CAM software, Oun 24):4

Schlumberger Technologies

HP Apollo 9000 Series 700 and Series 400 workstation agreement with, Oun 24):4 software revenue on technical workstations,

Oun 24):7

SCO

UNIX System V, Oul 8):8

SGI

CASEVision visual development environment,

(Aug 26):5

CodeVision toolset, (Aug 26):5 company. See Silicon Graphics Inc. nUS Explorer visual application environment,

(Aug 26):5

IRIS Indigo workstation, produa analysis of,

(Aug 26):4

IRISserver family of file servers

Data Station 2, (Aug 26):5

POWERfile 50, (Aug 26):5

POWERfile 100, (Aug 26):5

Siemens AG

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, Oul 8):3

Siemens Nixdorf Informationsysteme AG

Cadence EDA joint venture with, (Aug 26):l4

Silicon Graphics Inc.

CAD/CAM/CAE revenue of, Oul 8):3 market share, CAD/CAM/CAE, Oul 8):3 strategy of, (Aug 26):6

©1991 Dataquest Incorporated / 1290 Bidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011617

CAD/CAM/CAE

Software demonstrations

cost of, (Aug 2 0 : 8

Standards

IEEE standard 1076 (VHDL), 0un 24):15

Structural Dynamics Research Corp. software revenue on technical workstations,

Oun 24):7

Sun company. See Sun Microsystems Inc.

Network Extensible Window System, (Aug 20:10

Newsprint software, (Aug 20:10

SPARCprinter, (Aug 20:10

SPARCserver family, (Aug 20:10

SPARCstation

C:ADDS upgrade for, (Jul 8):4

HP MCAD software ported to, Qui 8): 10 performance of, (Jun 24): 5

SPARCstation IPC, (Aug 26): 10

SPARCstation SLC, (Aug 20:10

Sun Microsystems Inc.

CAD/CAM/C:AE revenue of, QM\ 8):3 company analysis of, (Aug 20:9 document imaging imaging partners of, (Aug 26):11 installed base by industry, (Aug 26):12 product line of, (Aug 26): 10 strategy for, (Aug 26): 10 market share, CAD/CAWCIAE, (Jul 8):3 restructuring of, (Aug 26): 11

SunSoft Inc. charter of, (Aug 20:10

SunTech Enterprises Inc. charter of, (Aug 20:10

Supercomputers

trends for, (Jun 24):2

UNIX

application software on CD-ROM, (Aug 20:7

IBM ADC, 0ul 22):2

SCO UNIX System V, (Jul 8):8

Valid Logic Systems Inc.

CAD/CMA/CAE revenue of, (Jul 8):3

EDA framework market of, (Aug 26):3 market share, CAD/CAM/CAE, (Jul 8):3 ports to SGI IRIS Indigo workstation, (Aug 26):7

Veiilog Hardware Description language (HDL),

Cun 24):15

VHDL International Inc. consortium, Qua 24):15

VHDL Users Group (VUG), Oun 24):15

VHDL (VHSIC Hardware Description Language),

Oun 24):15

VHSIC Hardware Description Language (VHDL),

Oun 24):15

Video technology

in Digital DECpc 433T Workstation, Oul 22):11

Viewlogic Systems Inc.

EDA framework market of, (Aug 20:4

VUG (VHDL Users Group), Oun 24):15

Technical workstations

software revenue on, (Jun 24):7 worldwide CAD/C:AM/CAE market for, QvA 8):2

Technology analysis

EDA with HDL-based design reuse, 0 " ^ 24): 12 object-oriented databases (OODBs), (Jun 24):8

Windows, Oul 8):7

Terminals. See Host-based terminals; Host-dep>endent terminals

W

Workgroup Technologies

Sun's imaging software partnership with,

(Aug 20:12

Workstations

Digital DECpc 433T Workstation, Oul 22):10 trends for, Oun 24):2

See also Technical workstations

Worldwide CAD/CAM/CAE industry

market portfolio for, Oul 8):2 trends for, Oun 24):2

Worldwide computer industry

trends for, Oun 24):2

U

United States

CAD/CAM/CAE market in, Oul 8):3

See also North America

X Window applications

Highland Software's Software Store, (Aug 2 0 : 8

Xerox Imaging Systems

Sun's imaging software parmership with,

(Aug 20:12

i

I

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011617

Dataoyest

n i n a company of

M M i l The Dun &Brads(iccl Corporation

Dataquest

Perspective

CAD/CAM/CAE

Vol. 1, No. 5

October 7, 1991

Market An.nlvsis

C4n/CAM/CAE Forecast Update

Mirroring the economy's current economic recession, the CAD/CAM/CAE market continues in the doldrums, showing a relatively flat growth pattern through the first half of 1991 compared with the first half of 1990. This article looks at the CAD/CAM/CAE industry service's updated forecast for the worldwide CAD industry and for each market segment.

By CAD/CAM/CAE industry staff Page 2

McDonneU Douglas to Sell Off Systems Integration Unit to EDS

When McDonnell Douglas Corporation (MDC) signed a letter of intent to sell off its CAD/

CAM subsidiary, McDormell Douglas Systems Integration Company, to Electronic Data Systems ((EDS), MDC disclosed plans that would effectively signal the company's exit from the

CAD business. As it stands, the proposed sale of MDSI to systems integrator EDS, w^hich is a subsidiary of General Motors (Corporation, also promises to have an impact on the automotive world if MDC's Unigraphics CAD applications software should fall under the control of EDS.

By Kathryn Hale and Michael J. Seely Page 5

SunSoft Solaris Sets Its Sights on Intel's Solar System

In September, Sun Microsystems' software subsidiary SunSoft unveiled its Solaris operating environment software ambitiously aimed at an expansion of Sun's horizons to include computers based on the Intel 80x86 microprocessor architecture. This article looks at the

SunOS-based UNIX operating environment offering of Sun's new subsidiary and examines its prospects in a market that pits it against the Advanced Computing Environment (ACE) initiative and alliances such as the recent IBM/Apple partnership.

By John DeArmon, Peter Francis, Kathryn Hale, and Patrick Waurzyniak Page 7

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE

Market Analysis

CAD/CAM/CAE Forecast

Update

Proving no more immune to the current recession than the economy as a •whole, a significant portion of the computer-aided design, manufacturing, and engineering (CAD/CAM/CAE) market turned in a relatively sluggish performance during the first half of 1991 compared with the first half of 1990. Financial results comparing first-half 1991 results for a select group of publicly held companies in the electronic design automation (EDA) and mechanical application areas showed a definite decline in revenue. As a result of this decline and other factors, the final compound annual gro^wth rate (CAGR) for total revenue during the forecast period,

1990 to 1995, has been lowered slightly to

11.7 percent from our previous forecast of

13.0 percent. In addition, some growth that was forecast for 1991 has been pushed back to

1992. Table 1 shows the final forecast for the total woridwide CAD/CAM/CAE market and by each application segment; Figure 1 illustrates those results.

The CAD/CAM/CAE market's current state is mainly due to the continued recessionary economic pressures throughout North America.

This forecast update is based on the final CAD/

CAM/CAE market share numbers published in

Table 1

CAD/CAM/CAE Worldwide Revenue Forecast

1 9 9 0

All Applications

Revenue (U.S.$M)

Systems

"Workstations

Mechanical

Revenue (U.S.$M)

Systems

Workstations

AEG

Revenue (U.S.$M)

Systems

Workstations

GIS/Mapping

Revenue (U.S.$M)

Systems

Workstations

14,365

488,639

525,125

7,449

224,878

251,290

2,095

117,356

121,213

1,653

50,693

53,959

Eleoronic CAE

Revenue (U.S.$M)

Systems

Workstations

1,688

60,916

61,735

10 Layout

Revenue (U.S.$M)

Systems

Workstations

PCB/Hybrid/MCM

Revenue (U.S.$]V1)

Systems

Workstations

454

9,280

9,295

1,026

25,517

27.6.33

Source: Dataquest (October 1991)

1 9 9 1

16,000

575,800

609,050

8,090

258,810

282,730

2,409

139,560

142,980

2,116

65,910

69,040

1,829

70,070

70,800

484

12,650

12,660

1,072

28,800

30.8.30

1 9 9 2

17,946

671,520

701,430

8,686

291,480

313,300

2,749

163.600

166,450

2,651

83,860

86,750

2,182

83,450

84,020

547

16,650

16,650

1,131

32,480

34.260

1 9 9 3

20,035

766,440

792,160

9,180

320,850

340,000

3,145

190,430

192,630

3,294

103,950

106,460

2,605

94,510

94,920

625

20,580

20,580

1,186

36,120

37,590

1 9 9 4

22,361

856,160

878,530

9,706

343,310

360,610

3,571

214,290

215,970

4,069

124,360

126,420

3,050

107,880

108,180

714

24,620

24,620

1,252

41,690

42.740

4,079

237,270

238,570

4,964

141,600

143,220

3,638

121,780

121,990

811

29,490

29,490

1,351

47,340

48.040

1 9 9 5

24,964

931,960

951,100

10,121

354,490

369,800

CAGR (%)

1 9 9 0 - 1 9 9 5

11.7

13.8

12.6

6.3

9.5

8.0

14.2

15.1

14.5

24.6

22.8

21.6

16.6

14.9

14.6

12.3

26.0

26.0

5.7

13.2

11.7

©1991 Dataquest Incorporated / 1290 Ridder Parle Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE

Figure 1

CAD/CAM/CAE Forecast by Application

Millions of Dollars

14000

• Mechanics^

12000

• AEC

X QIS/MappIng

10000

6000

6000-1

4000

2000 j ^

0?

1990 1991

Source: Dataquest (October 1991)

A Electronic CAE

+ IC Layout

• PCB/Hybrld/MCM tasit

June. (The 1990 market totals will match in both the market share tables and forecast tables). Any changes in market conditions have been incorporated into this final forecast, and updated forecast tables are currently being published in the new Source: Dataquest format.

The following analysis outlines the reasons for any changes to the CAD/CAM/CAE market forecast that occurred between March and October

1991.

Dataquest Perspective

The CAD/CAM/CAE market's current state is mainly due to the continued recessionary economic pressures throughout North America, but it is also due to the fact that many companies have slowed their capital equipment expenditure, delaying some purchase decisions in the face of weak economic conditions.

Applications

Mechanical Applications

The mechanical CAD/CAM/CAE market performed about as expected on a worldwide basis compared with our forecast a year ago. This high-level view hides some of the underlying turmoil incumbent in the mechanical applications industry. The CAGR for revenue from

1990 to 1995 has been lowered to 6.3 percent from 9.0 percent. The personal computer-based mechanical products grew^ faster than the forecast revenue growth of 14 {jercent, reaching an

18 percent level growth for 1990, as recession-

^m^ wm

1995 ary pressures in the United States and abroad focused more attention on this lower-cost solution.

Host-based mechanical products did significantly better than expected, growing at a 6 percent rate rather than having the slight decline that was forecast for the platform. The primary reason for this change is the continued expansion of host-based products in the Japanese market, as many of the large Japanese CAD/CAM suppliers aggressively sell new and add-on seats into the growing markets of the automotive and electronics industries.

Although there is no recession in

France, Germany, or Italy, the troubled economies in the United

Kingdom and the Scandinavian countries are recovering somewhat slower than was expected in 1990.

Technical workstation-based products have had exceptional growth rates in the last few years but now are slowing down to a more reasonable 20 percent growth rate, thus experiencing slower-than-expected revenue growth. The technical workstations CAGR revenue forecast for

1990 through 1995 has dropped from 17 to

13 percent, and workstation revenue growth for 1991 dropped to 18 percent.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE

Some of this revenue decline in technical workstations is due to introductions of many outstanding price/performance packages and very aggressive pricing on older models. Although unit growth for the technical workstation-based mechanical products is somewhat off-pace,

Dataquest's forecast remains aggressive over the long term, with unit volume projected to double in the next five years. Prices of technical workstation products are expected to continue to erode as the desktop market focuses on the key competitive batdefront of the engineer's desktop.

AEC

The architecture, engineering, and construction (AEC) market forecast remains relatively unchanged and is still expected to grow^

15 percent in 1991. The relatively low CAD penetration rate in AEC, combined w^ith client demand to produce drawings in electronic form, encourages both large and small A/E enterprises to continue investing in new CAD systems despite the difficult economic environment.

GIS/Mapping

The geographical information systems (CIS) forecast is also relatively unchanged since

March and is exf>ected to grow 28 percent in

1991 with a 25 percent CAGR through 1995.

Historically the fastest-growing segment of the

CAD/CAM/CAE market, GIS is holding strong.

Despite recessionary pressure looming over the previously high-growth state and local government market in North America, new technologies (such as global positioning systems combined with improved aerial photography) fuel buyer interest in GIS systems, while new applications and sales sites (such as Kuwait) continue to unfold.

EDA

In the EDA markets, revenue projections have declined somewhat from preliminary 1991 forecasts that were made in March. The changes in our forecast for the EDA market are due to several factors that resulted in a 13.2 percent drop in first-half 1991 revenue for a select group of public EDA vendors.

Among the changes, we have revised our 1990 to 1995 revenue projections (including hardware, software, and service) for IC layout from a CAGR of 16 percent to 12.2 percent, and printed circuit board (PCB)/hybrid/multichip module (MCM) has dropped from a CAGR of

10 percent to 5.7 percent. The forecast for electronic CAE has remained relatively unchanged.

The primary reason for the overall slowdown in

EDA, particularly the PCB and IC layout markets, is the fundamental erosion of hightechnology profit margins together with a saturation of the PC market. We expect this trend to continue into 1992.

Dataquest believes the EDA market's current state can be traced to the following factors:

• Approximately 60 percent of the CAE software is sold into the ASIC design market, and more than 60 percent of the ASIC design market in North America is composed of computer manufacturers, which have been hit hard by the recession.

• In addition, the economy in general and the

Persian Gulf w^ar in particular had severe effects on the EDA industry, causing many

EDA customers' capital spending plans to be put on hold for the first half of the year.

• The Gulf war affected all parts of the world including Japan, which accounts for a 28.7 percent of the worldwide EDA software market, and this situation was further exacerbated by the recession in the United States.

• The fact that large EDA vendors still have not been able to meet the market's product requirement demands also is adding to the market slowdown.

• Despite the faa that large companies are not gaining ground, some smaller niche EDA companies such as Synopsys Inc. and

Viewlogic Systems Inc. have actually done well, offsetting some of the setbacks suffered by large EDA vendors.

• Softness in the semiconductor industry has also adversely affected the IC layout market.

The U.S. semiconductor industry's book-to-bill ratio has been down each month for the past three months, reflecting both the semiconductor industry's traditional summer doldrums and the general economic malaise that has afflicted electronics manufacturers.

The Regions

Europe

Overall, Europe appears to be on target with our forecast, with no change to the European market's CAGR while posting slightly higherthan-expeaed growth. Although there is no recession in France, Germany, or Italy, the troubled economies in the United Kingdom and the

Scandinavian countries are recovering somewhat slower than was expected in 1990.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE

On the plus side for European electronics manufacturers is the expectation that, in spite of the termination of the Cold War, defense equipment expenditure over next two years is unlikely to be reduced. Because equipment must to be replaced now that the Persian Gulf war is over, and the Europ>ean countries' armed forces continue to trade personnel (soldiers) for technology (such as smart bombs, communications equipment, and missiles), continued strong spending on defense will help the United Kingdom, France, Italy, and Germany to maintain currently projected growth rates.

Asia

The mechanical and EDA forecasts in Asia both have dropped significantly, primarily because of the slower sales of products for the technical workstation platform. Growth rates for Asia in

1991 are projected to be 24 percent, down from the 31 percent forecast last year. The

CAGR also has fallen from 14 to 12 percent, making the growth projections for Asia lower than those for Europe.

Japan continues to provide the emerging EDA opportunity, although end users there are waiting for the arrival of open systems before aggressively snapping up large quantities of

EDA tools. We believe that open system "standards" will begin emerging in 1992.

Overall, capital investment in Japan grew at a

7.3 percent annual rate during the April to June quarter compared with a year earlier, including capital investment by manufacturing companies rising 18.8 percent during the quarter. In addition, Taiwan's economy also is showing strong signs of recovery, with at least 7 percent growth expected for this year. The Taiwanese economy's growth is attributable primarily to booming exports, which surged to 12.5 percent during the first six months of 1991 versus the first half of 1990 and rising domestic investment spurred by tax incentives and promoting the privatization of 20 state-run companies.

Forecast Methodology

Fundamental to the way Dataquest conducts its research is an underlying philosophy that says the best data and analyses come from a well-balanced program. This includes balance between primary and secondary collection techniques; balance between supply-side and demand-side analysis; balance between focused, industry-specific research and coordinated, "bigpicture" analysis aided by integration of data from the more than 25 separate high-technology industries Dataquest covers; and balance between the perspectives of experienced industry professionals and rigorous, disciplined techniques of seasoned market researchers.

The CAD/CAM/CAE industry market estimates and forecasts are derived using the following research techniques:

• Bottom-up aggregation—^This method involves adding all relevant vendor contributions to arrive at total market estimates for all historical data.

• Segment forecasting—For each application segment tracked by the CAD/CAM/CAE industry service, individual forecasts are derived following a basic information model.

Specifically, each design phase covered within each application is segmented by channel, product, region, and platform. In this way, each application segment incorporates its own set of unique assumptions.

• Demand-based analysis—Market growth is tracked and forecast in terms of the current and anticipated demand of current and future users. This analysis requires the development of a total available maiket model and a satisfied available market figure to accurately assess the levels of penetration. Installed base and rates of product retirement are also evaluated. In addition, Dataquest analysts factor in the acceptance or ability for users to consume new technology.

• Capacity-based analysis—^This method involves identifying future shipment volume constraints. These constraints, or "ceilings," can h>e the result of component availability, manufacturing capacity, or distribution capacity. In any case, constraint is capable of keeping shipments below the demand level. •

By CAD/CAM/CAE industry staff

©1991 Dataquest Incorporated / 1290 Rldder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE

McDofmell Douglas to

SeU Off Systems

Integration Unit to EDS

In August, McDonnell Douglas Corporation

(MDC) and Electronic Data Systems Corporation

(EDS) signed a letter of intent under -which

EDS will acquire MDC's computer-aided design, manufacturing, and engineering (CAD/CAM/CAE) subsidiary, McDonnell Douglas Systems Integration Company (MDSI), for an undisclosed sum.

The proposed sale, which was expected for some time, is subject to several conditions and the negotiation of agreements covering the transaction. The companies said that the deal would include all of the domestic operations of

MDSI and certain CAD/CAM/CAE research and distribution activities of both MDSI and McDonnell Douglas Information Systems International outside of North America.

Timing of the sale of MDSI is not a surprise considering the recent debt-ridden history of MDC's airplane business.

Although no financial terms of the proposed deal were disclosed, outside sources estimated that the sale price potentially could range upward of $225 million to even as much as

$350 million. A sale price reaching the high end of that scale undoubtedly would help alleviate the strain of MDC's sagging balance sheet, which as of the second quarter included

$2.94 billion in debt.

EDS, a leading systems integration and information technology company that has 64,000 employees worldwide, coveted the MDC subsidiary's Unigraphics software applications, which holds a strong position among such customers as Advanced Microelectronics Products

Inc. (AMD), General Motors Corporation (GM),

MDC, and Pratt & Whitney. During 1990, EDS had worldwide revenue of $6.1 billion.

MDSI, which serves the industries of insurance, manufacturing, infrastructure, telecommunications, and government, reported worldwide revenue of $398 million in 1990 and an increase in earnings over 1989.

Dataquest Perspective

The "For Sale" sign is firmly planted outside the MDSI complex. If the current chain of events continues, it may not be there long. EDS is aaively pursuing this business deal. Dataquest believes that EDS is acting on a "right-offirst-refusal" clause formed in the award contraa when MDSI was selected more than two years ago as a primary vendor to GM's C4 program. The clause probably was included as a defensive measure rather than planned as an offensive opportunity, although the combined business opportunity is the current focus of discussion.

The strong financial position of EDS is fueling interest in an ever-Tvidening industry base. The

July 1991 agreement with Consilium Inc., a leading software developer in plant floor management applications, indicates a growing interest by EDS in supplying computerintegrated manufacturing systems development, systems integration, and management services.

The plan to acquire MDSI fits well into this strategy. MDSI is well respected for its products in the CAM application area. Existing relationships between MDSI and ICAD Inc., PDA

Engineering, and Valisys Corporation would further enhance the position of the EDS as a systems integrator in a broad range of CAD/

CAM/CAE applications.

EDS and MDC are expected to announce the details of the final agreement soon. Assuming completion of the deal, the MDSI employees involved are expected to become EDS employees with new orders and strategic incentives. John Mazzola, currentiy president of MDSI operations, has armounced his intention to stay involved with the new venture. He will report to Hank Johnson of EDS and be responsible for the Unigraphics business unit.

Timing of the sale of MDSI is not a surprise considering the recent debt-ridden history of

MDC's airplane business. Reduced defense spending, cost overrun problems with the A-12 and C-17 programs, and scheduling problems with the ne-w wde-body MD-11 are forcing a very careful look at the bottom line. An estimated $250 million to $350 million deposit in

MDC's account after the sale would be welcome. Rumored inquiries from Fujitsu Limited,

Siemens, and perhaps others, apparently were less attractive because of the cash interest by

MDC. The suspected joint-development or marketing agreements sought by such companies would not have been as valuable to MDC in the short term. The existing distribution agreement with Fujitsu for the Unigraphics produas

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE in Japan has been in effect since January, and under an earlier agreement with MDSI, Seiko

Instruments Inc. is distributing Unigraphics on other platforms.

Dataquest's estimates that MDSI's CAD/CAM/CAE worldwide 1990 revenue was $352.7 million, a

25.1 percent increase over the McDonnell

Douglas MDC unit's estimated 1989 revenue of

$282 million. During 1990, Dataquest research indicated that MDSI gained its highest and strongest growth in the architecture, engineering, and construction (AEC) and geographical information systems (GIS) application segments.

According to Dataquest's estimates, these areas showed growth rates of 399 and 37.2 percent, respectively, while MDSI increased mechanical applications growth by 20.1 percent. An estimated $15 million in earnings were reported for 1990 by MDSI.

EDS had a good year in 1990 as well. EDS reported a record $6.11 billion in revenue, a

12 percent increase over 1989, which resulted in net income of $497 million. In addition, over the past 10 years, EDS' revenue has grown by a factor of 20, while the company increased its head count from about 10,000 employees to over 61,000 worldwide.

The MDSI agreement is one of several recent purchases and equity positions taken by EDS.

(Other examples include Com Systems Inc., of

Westlake Village, California; Capsco Software of

Canada; Hitachi Data Systems Corporation of

Santa Clara, California; and ASK Computer Systems Inc. of Mountain View, California.) EDS is enormously successful as a systems integrator and has the money to spend to create an increasingly larger power base.

If consummated, the deal will give EDS full control over a major software vendor installed as a primary vendor in the GM C4 program.

This deal also opens the door for EDS to pursue the installed base outside GM, where

MDSI has been very successful in selling manufacturing-oriented systems to Fortune 1000 companies. The team of EDS service organization and consultants, combined with the MDSI group of trained system engineers, potentially makes a formidable team with the potential for a strong product engineering to manufacturing automation scenario. And the bottom line for the MDSI group would be gaining a parent company that sees more direct value in supporting the CAD/CAM/CAE and GIS business.

From the perspective of MDSI's GDS product line of AEC and GIS software, the quality of ownership just took a big jump upward. The

GDS line had no obvious role in MDC's enormous aerospace business. As a subsidiary of the largest information systems integrator, the GDS sales force is likely to find more prospects, and the product's competitive position can be expected to improve. EDS' broad commercial and government systems integration business will benefit from owning a proven GIS and

AEC line clearly positioned as an "infrastructure life-cycle management" product.

Delicate areas are likely to surface, however.

EDS Canada and Intergraph Corporation both recently helped finalize a significant land records consortium for the province of Ontario,

Canada—a consortium that will break new ground in distribution of data modeled as a public utility. Considerable future opportunity will be created if this model is successful— opportunity that could compete with the GDS line. EDS must maintain a reputation for independence when proposing software that it already owns. •

By Kathryn Hale

Michael J. Seely

SunSoft Solaris Sets Its

Sights on Intel's Solar

System

In a move to expand into the vast Intel territory, Sun Microsystems Inc.'s software subsidiary

SunSoft Inc. last month announced a new direction with what it termed the computer industry's first "shrink-wrapped" distributed soft^svare operating environment. Dubbed Solaris, the Sun-

Soft operating environment is aimed not only at

Sun's SPARC-based RISC workstations but also squarely at the much larger installed base of personal computers based on Intel Corporation's microprocessor architecture.

Based on Sun Microsystems' 32-bit UNIX operating system SunOS, Solaris 2.0 is targeted at

SPARC-based RISC platforms as well as the many thousands of PCs based on the ubiquitous Intel 80x86 architecture. However, the

Solaris operating environment is likely to run only on the fastest of PCs incorporating either

Intel's 33-MHz version of the 80386 chip or the much newer 80486 microprocessors.

SunSoft's introduction of Solaris, made before an audience of independent software vendors

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE

(ISVs) attending SunSoft's Catalyst Developers

Conference, is the company's first major move since Sun spun off the system softw^are subsidiary last February. SunSoft touted Solaris as the first operating environment with symmetric multiprocessing and multithreading w^ith application-accessible threads, enabling highperformance commercial applications such as on-line transaction processing and database management.

Figure 1

Solaris 2.0: The Industry's Shrlnk-Wrapped

Distributed Computing Solution

Although Sun has held the mantle of market leadership in the fastgrouHng workstation field for quite some time, the coveted goal of branching out into the huge commercial business applications market had remained elusive.

Divided into three layers, the Solaris environment consists of the SunOS 5.0 operating system, which incorporates UNIX System V Release

4 (SVR4) and the Open Network Computing

CONC) networking standard (see Figure 1). The application framework includes a new developer environment called OpenWindows Version 3, which gives developers the newly announced

SunSoft ToolTalk object-oriented application interof)erability solution and other develojser tools with which ISVs can integrate programs into the network.

On the end-user side, SunSoft claimed that

Solaris places network resources at users' fingertips through both an intuitive 3-D Open Look interface desktop metaphor and the company's

DeskSet suite of 15 productivity tools, work group, and multimedia applications. Solaris 2.0 has international support for customization worldwide according to local language requirements.

Distributed Ob|ect-Oriented

Applications

Another advancement of Solaris is the use of an object-oriented scheme that SunSoft calls

Project DOE (DOE stands for distributed objects everywhere), under w^hich SunSoft hopes to commercialize object-based computing into the computer industry mainstream. Unlike personal computer operating systems requiring total rewrites to move objects, SunSoft's Solaris 2.0 purports to provide a smooth path to build on the SunOS foundation. Using the object-oriented technology of Project DOE, which is based on

Source: Sun Microsystems Inc. technology that was jointly submitted with the

Hewlett-Packard Company to the Object

Management Group, SunSoft aims to gready reduce the complexity of building sophisticated, distributed applications that can interoperate across multiple platforms.

Solaris SunOS 5.0 complies with industry standards and sjjecifications including IEEE POSIX

1003.1, X/OPEN Portability Guide (XPG), ISO

9660 AT&T System V Interface Definition (SVID)

3, and the SPARC Compliance Definition (SCD)

2.0. SunSoft maintains that support of SCD 2.0 means software developers will be able to create shrink-wrapped applications that w l l run on all SPARC-based UNIX systems without modification.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE

In addition, the combination of Sun's SunOS

5.0, ONC, OpenWindows, Open Look, and

DeskSet are integrated into a single package that SunSoft said will run on both Sun SPARC

RISC platforms and the Intel 80x86 platforms.

SunSoft said software written for Solaris 1.0, the company's distributed computing environment based on SunOS 4.1.1, is source-compatible with Solaris 2.0. The latest Solaris version also is source-code compatible with the Intel and

SPARC platforms, thus providing the largest compatible application base in the 32-bit market encompassing more than 3,600 third-party software and hardware solutions.

SunTech To Provide CASE/Development Tools

Another recently formed Sun subsidiary, Sun

Technology Enterprises, or SunTech, will build and market tools for the development of SCDcompatible applications. The products to be marketed by the new company include SCDcompatible SPARCprinter, NeWSprint software,

PC-NFS, SPARCompiler software development platforms, and SunLink connectivity gateways to multivendor systems and wide-area networks.

Dataquest expects these tools to greatly increase the amount of applications available on the

SCD-compatible architectures in the next two to five years.

CD-ROM and documentation. Pricing is $1,395 for traditionally configured workstations and

$795 for low-cost commercial systems such as laptops. Solaris 2.0, based on SunSoft's

SVR4-based SunOS 5.0 with symmetric multiprocessing and multithreading plus enhanced

ONC, OpenWindows, and DeskSet, will be offered for both Sun SPARC and Intel platforms and is available in developer copies now through SunSoft. Early access release of Solaris

2.0 for both SPARC and Intel machines is scheduled for early 1992, while volume shipments are planned for both platforms during the first half of 1992, at which time SunSoft will announce pricing for the produas.

Vendor Support

The SunSoft Solaris armouncement immediately garnered much third-party support from software and hardware vendors alike. More than 50 software developers spanning the commercial, productivity, and technical markets thre'w their support behind the SunSoft environment. Four hardware companies endorsing Solaris included

AST Research Inc., CompuAdd Corporation, Dell

Computer Corporation, and Toshiba Corporation.

SunSoft's Strategy likely will make the company much, much more successful than when it was just

Syn Microsystems, the hardware company.

Now, Sun Microsystems has clarified the battle lines, facing off against ACE and the IBM/Apple camp in what has become the highest-stakes power game in the computing business.

The SunTech subsidiary initially is repackaging

Sun's existing software tools and providing them as unbundled value-added products to the development community. Sun traditionally provided these tools at little or no cost to developers and end users with the expeaation of increased hardware sales. As hardware margins have decreased and developers have expressed the desire for more sophisticated development tools and environments. Sun views this service as an additional business opportunity.

SunSoft will offer two versions of Solaris. The first version, Solaris 1.0, based on SunOS 4.1.1,

OpenWindows Version 2, and DeskSet Version

2, is available immediately to all SPARC vendors in a shrink-wrapped package that includes a

SunSoft also enlisted support from Novell Inc., which signed on to distribute Solaris through the Novell reseller channel as well as to incorporate Novell NetWare network operating system products into the new distributed operating environment software. In addition, ASCII

Corporation, Japan's largest PC and UNIX software company, announced it wiU distribute

Solaris through its resellers in Japan, and ICL, the leading computer maker in Europe, will collaborate with SunSoft to enable SVR4 applications to run unmodified on multiple SPARC platforms.

Dataquest Perspective

Ever since early April, when nearly two dozen computer industry heavyweights such as Digital

Equipment Corporation, Microsoft Corporation, and Compaq Computer Corporation banded together to throw their collective weight behind

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

10 CAD/CAM/CAE the formation of the Advanced Computing

Initiative (ACE), it was painfully obvious that

Sun Microsystems would have to do something serious to strike back at the ACE consortium promoting a computer platform based on the

RISC microprocessors of Sun's rival, MIPS Computer Systems Inc.

Sun Grows Up, Targets Commercial

Applications

Although Sun has held the mantle of market leadership in the fast-growing workstation field for quite some time, the coveted goal of branching out into the huge commercial business applications market had remained elusive for the Mountain View^, California-based workstation company. Indeed, for some time now,

Sun has been desperate to migrate its systems from the technical platforms it dominates toward the much larger market of commercial computers.

Sun's original strategy was heavily entrenched in an orientation skewed toward not only hardware but also the technically oriented marketplace. The SunSoft spin-off, however, is a significant and promising departure from Sun's previous approach, one that holds considerable potential to control more desktops. As hardware price/performance drops, vendors that control the software tools, frameworks, environments, and applications will be capable of playing a dominant role in the future computer industry.

Therefore, SunSoft's strategy creates an opportunity for Sun to become more successful than when it was just Sun Microsystems, the hardware company. Other oi^anizations are now doing similar things. For instance, Xerox Corporation recently arvnounced a software unit. Part of the rationale for spinning off the software unit is that such operations can be much faster alone and also capable of culturally disentangling itself from the parent oiganization. In the past, hardware vendors have had competing strategies leaning toward the hardw^are side of the business, exacerbating the challenges in providing adequate software development tools and environments. By concentrating their software efforts within a separate organization, hardware companies stand a better chance of competing in the increasingly more profitable software business.

Strong intrinsic differences exist between the needs of business and technical applications.

The platforms on which these technical and business applications are created and deployed, however, reflect primarily de facto standards, such as UNIX in the technical world or DOS and MVS for business applications. Because the platform differences that exist between the market segments reflect primarily de facto standards, UNIX platforms can be successfully marketed and utilized in the business applications arena.

A major problem with UNIX in general is that it has always been large and cryptic, demanding more development, computing, and maintenance resources than DOS-based systems, thus making it less practical for small business settings. If business applications were more readily available on UNIX platforms, those UNIX platforms would be a viable alternative for today's business computing needs. A prerequisite for creating UNIX business applications is the availability of development tools, environments, and frameworks, which provide facilities for the intrinsic needs of those business systems. Sun has taken a lead in providing a high-level operating environment on the Intel platform, the predominant desktop business platform, giving it an advantage in creating a standardized, integrated application environment for developing advanced UNIX-based business systems frameworks.

With SunSoft's attempt to branch out Sun's reach from solely UNIX-based SPARC platforms into the gigantic market posed by the installed based of millions of Intel-based PCs, which represents roughly 80 percent of the installed base of desktop computers. Sun in effect appears to be saying, "He who has the most applications wins." Dataquest believes that Sun's bold step with its SunSoft Solaris operating system for Intel- and SPARC-based computers is a significant move toward remedying Sun's problems breaking into business-oriented markets.

Desktop Wars and Mega-Alliances

With the formation of ACE, and the more recent advent of IBM Corporation's collaboration with Apple Computer Inc. on its Pink operating system software, the trend today seems to be tow^ard big-company alliances each staking out consortia claims on the lucrative desktop computing market. Now^, Sun Microsystems has clarified the battle lines, facing off against ACE and the IBM/Apple camp in what has become the highest-stakes p)ower game in the computing business.

The alignment of Sun with chip vendor Intel might seem unusual, given Sun's steadfast promotion of its SPARC RISC microprocessor, until

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

CAD/CAM/CAE 11 one considers the sheer volume of the installed base represented by Intel's 80x86 platform.

Although Solaris is meant only for the fastest of

Intel 386- and 486-based PCs, by the time the operating environment reaches the market Intel is likely to have early versions of its 80586 microprocessor available, as well as faster versions of the 486 chip in 40- and 50-MHz speeds. One faa stands out in the desktopconsortia -wars—^it has become ever-more important for computer vendors not only to align themselves with partners but also to team up with the right ones. Until Sun picked Intel's platform for Solaris, Intel and its overpowering installed base had been overlooked in the latest round of partnering announcements. Sun correctly tapped Intel as a key player in the computer industry.

Solaris: A Safe Port for ISVs?

While Sun's reputation has been made as a hardw^are vendor rather than a software-oriented company, SunSoft and Sun's recently created

SunTech software tools subsidiary should make headway toward changing that perception. In fact, Sun's move toward the mainstream commercial market bodes well for the many ISVs interested in writing applications for Solaris on

SPARC- and Intel-based machines. Although Sun is so large that it seems it cannot always afford to help out small companies that get lost in the shuffle, many ISVs undoubtedly will recognize this market opportunity as a safe jxDrt given

Sun's position as a very stable company and the workstation market share leader. With

Solaris, SunSoft seems to have the beginning of an edge of the ACE camp.

In addition, on September 26, Sun Microsystems and SunSoft announced that SunSoft would acquire the Intel-UNIX of)erating systems software operations of Interactive Systems

Corporation, a subsidiary of the Eastman Kodak

Company based in Santa Monica, California, which reportedly has worked with Sun in the past to port versions of its SunOS to various other hardware platforms. Interactive currently offers several software operating environments for 80386- and 80486-based PCs, including the

Interactive UNIX Operating System Release 4.0 and VP/ix, a UNIX and DOS-compatible multiuser, multitasking operating system.

The viability, performance, and quality of an operating system is a fundamental component of any system software applications. Under the agreement between Sun, SunSoft, Interactive, and Kodak, SunSoft will acquire Interactive's

Systems Product division, which develops software and markets for the Intel platform, thus giving SunSoft an immediate market presence on the Intel 80x86 architecture with Interactive's product and market position. The companies also announced Sun's intention to license certain unspecified software technologies from

Kodak. Terms of the acquisition and the technology licensing agreement were not released.

In summary, SunSoft's Solaris announcement overall is a very positive and reassuring move by Sun Microsystems to provide a distributed runtime environment that is consistent across multiple platforms. The major hardware vendors have been attempting to provide such uniform and consistent runtime environments across multiple, interoperable platforms, although to date they have not been successful in this ambitious endeavor. Sun's strategy is an aggressive, bold move designed to provide a truly value-added product to the end-user community. •

By John DeArmon

Peter Francis

Kathryn Hale

Patrick Waurzyniak

(This article also appears in the Personal Computer Software Dataquest Perspective.)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

12

In Future Issues

Look for the following topics of interest to the

CAD/CAM/CAE industry in upcoming issues of

Dataquest Perspective:

• A report on electronic design automation

(EDA) applications

• An overview of the architecture, engineering, and construction (AEC) industry

• More "News and Views"

CAD/CAM/CAE

For More Information . . .

Managing Editor Patrick Waurzyniak (408) 437-8132

On the topics in this issue CAD/CAM/CAE (408) 437-8132

On related subjects Client Inquiry Center (408) 437-8671

About other Dataquest publications Sales (408) 437-8000

About upcoming Dataquest conferences Conferences (408) 437-8245

About your subscription Customer Service (408) 437-8402

Via fax request Fax (408) 437-0292

The content of this report represent our Interpretation and analysis of information generally available to the public or released by responslUe individuab in the subject companies, but Is not guaranteed as to accuracy or completeness, h does not contain material fHovided to us in confidence by our clients. Individual companies reponed on and analyzed by

Dataquest may be clients of this and/or other Dataquest services. This information is not furnished in connection with a sale or offer to sell securities or in connection with the solicitation of an offer to buy securities. This firm and its parent and/or their officers, stockholders, or members of their flunllies may, from time to time, have a long or short position in the securities mentioned and may sell or buy such securities.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011474

Dataqyest

3 company of

The Dun & Bradslrcct Corporation

Dataquest

Perspective

CAD/CAM/CAE

Vol. 1, No. 4

August 26, 1991

Market Aritilysrs

Impending Framework Fragmentation Triggers New EDA Business Strategies

Contrary to electronic design automation (EDA) users' expectations, a single, homogenous, and open framework for the software tools used in produa design is not likely to happen soon, if at all. The following Dataquest research examines the impending fragmentation of the EDA framework market and its impact on the business strategies of EDA vendors.

By Ron Collett

Page 2

Silicon Graphics Announces Indigo, a New Low-End Workstation for Less than $10,000

With the latest workstation from Silicon Graphics Inc. (SGI), Indigo, SGI effectively brought the entry point of 2-D and 3-D graphics down to a level where sophisticated 3-D modeling may begin to reach the masses. In addition, the SGI workstation immediately gained support from the CAD/CAM/CAE market with EDA vendors Cadence Design Systems and

Valid Logic Systems announcing ports to the machine, opening up the prospect of future

3-D modeling of chip design for the EDA market.

By Laura Segervall and Michael J. Seely Page 4

Highland Software Ships Software Store

Highland Software is shipping the Software Store, a compendium of UNIX application software on a CD-ROM disc that is distributed at no cost to qualified UNIX workstation users for evaluation. Offering independent software vendors the chance to demonstrate their

UNIX software to prospective buyers, the Highland Software program currently distributes three versions aimed at users of UNIX software on workstations from the Hewlett-Packard

Company, Sun Microsystems Inc., and Digital Equipment Corporation.

By Gladys Francis and Kathryn Hale Page 7

Coinnany Analysis

Workstation Vendor Strategies in the DIMS MarketSun Microsystems

Sun Microsystems Inc. reported document imaging revenue totaling $140 million for its 1991 fiscal year, a 40 percent growth over last year. The company is making successful inroads into the document image management systems (DIMS) market through persistent adherence to standards in product development, marketing, customer support, and channel cultivation.

This article examines Sun's imaging strategy and comments on the company's chances for long-term success in the DIMS market.

By Pamela Stone Bliss

Page 9

News jn(] Views

An analysis of recent news events affeaing the CAD/CAM/CAE industry

Cadence and Siemens Nixdorf Enter Joint EDA Venture

Page 14

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-SOOO / Fax (408) 437-0292

0011243

CAD/CAM/CAE

Market Analysis

Impending Framework

Fragmentation Triggers

New EDA Business

Strategies

Dataquest's research indicates that the EDA market has been somewhat preconditioned to believe that a single, homogenous, open framework for electronic design will be established in the short term. The market anticipates that this framework will emerge from the myriad of vendors marketing frameworks and the CAD

Framework Initiative (CFI).

Dataquest, however, does not expea this to occur for at least five years, if at all. Rather, we believe that the framework market will undergo considerable fragmentation over the next two to three years. In short, a stampede of different suppliers' frameworks will penetrate the EDA user base, leaving a fragmented market in its wake (see Figure 1). Although frameworks from different suppliers will have a different

"look-and-feel," most, if not all, will incorporate the standards ratified by the CFI. As a result, the major incompatibilities among frameworks that will emerge in the short term will diminish over time. The following research examines the dynamics of this splintered framework market and offers a perspeaive on what it means for

EDA suppliers in the future.

EDA Frameworks: A Lack of

Standards

In spite of the EDA market's need for standards, a salient fact is that few standards exist today. This is particularly true in the area of

EDA frameworks. A framework comprises a multiplicity of capabilities, but Dataquest's research indicates that both the industry and the market have varying views on exacdy what constitutes a framework. However, staying within the guidelines pronounced by the CAD

Framework Initiative, a framework comprises the following eight elements (for which CFI has formed technical committees):

• Architecture—^The interdependencies and interrelationships among functional areas of a framework

• System environment—Operating system services necessary to provide hardware independence to other subsystems within a framework

Figure 1

Framework Fragmentation

>

(D

_ l

c o

Mentor

Cadence

Valid -

Dazix-

View Logic

CAD Framework Initiative

The EDA

Framework Wars and Standards

CAD Framework Initiative

1991

1992

Source: Dataquest (August 1991)

1993 1994 1995 1996

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE

Design data management—The mechanisms for storing, accessing, and versioning of design data

Design methodology management—^The set of activities necessary to create and complete a design

Design representation—^The conceptual data models used to describe the various elements of a design

Intertool communication—^The mechanisms for efficient sharing of design information between tools

User interface—^The portion of the system that manages all interactions with the designer or other user

Component information representation—^The digital representation of elearical components including models, libraries, and distribution of component information

The big question that the market continues to ask is which framework uHll ivinF Dataquest believes that clear winners will not emerge for another three to five years.

There has been a proliferation of framework announcements from many different vendors, which is leading to a more intensified and protracted batde for framework market share.

Among the most aggressive vendors digging in for the impending framework wars are Mentor

Graphics (forporation, Cadence Design Systems

Inc., Valid Logic Systems Inc., Dazix/Intergraph,

Racal-Redac Inc., Viewlogic Systems Inc., and

Digital Equipment Corporation (DEC).

What does supplying the framework to the customer mean strategically for EDA vendors? Dataquest believes that it means account control— the vendor that supplies the framework will essentially "own" that customer. Account control more or less equates to an ongoing revenue stream from that particular customer. We anticipate that customers will look first to their framework supplier when investigating the purchase of additional products and services. Thus, for large EDA vendors whose businesses demand a high revenue stream, the framework is of enormous strategic importance. But the marketplace, we believe, will continue to exhibit near indifference to framework offerings in the absence of EDA application software that is of superior capabilities and integrated into the framework. In other words, the framework alone is of littie value without a robust suite of tools. Yet the framework itself will have to stand on its own merit. Therefore, EDA vendors must deliver both in order to establish a framework beachhead.

Dataquest believes that the winners in the framework "wars" tvill not necessarily have the best framework, but rather will be able to deliver the best EDA solution to the customer.

Dataquest expects the North American EDA framework market to be 60 to 70 percent penetrated by 1995, with Asia and Europe following by 1996. In our view, EDA vendors currently possessing the largest installed base will wield both a tactical and strategic advantage over competitors in the imj>ending framework battles. Of course, it is important to evaluate installed base within the context of account control—the value of a laige installed base is seriously diminished if account control is lacking within the base.

Using installed base as a proxy for competitive advantage. Mentor is at the top with a base of approximately 22,000 seats, following by Valid with approximately 15,000 seats, and Dazix/

Inteigraph with about 11,000 seats. Other contenders include Cadence, Viewlogic, and

Racal-Redac.

The market is not concerned with where the value is created, so long as value is delivered in an uninterrupted fashion.

In spite of Cadence's sheer size and the fact that it has high brand recognition in the framework arena, the only market where the company wields a significant edge is the IC layout sector. Here the company has approximately

4,000 seats, which correspond to approximately

36 percent market share. Yet in the systems market. Cadence's strength is tenuous. The company's presence has been limited almost exclusively to the Verilog simulator, and the Verilog simulator has traditionally resided with another

EDA vendor's environment. Thus, at best,

Cadence shares account ownership of its Verilog

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE installed base with compaiues such as Mentor and Valid. We believe that this puts Cadence at a significant disadvantage in the framework market share wars, although it is reasonable to assume that some of Cadence's power in the

IC layout market will spill over into the systems arena.

Racal-Redac is also at a disadvantage in that it, too, shares ownership of its customers with many other suppliers. Its installed base is split into two distinct camps—the HHB Systems CAE installed base and the Racal-Redac printed circuit board (PCS) layout base. Viewlogic faces similar challenges in the framework wars because the majority of its installed base (approximately 8,000 seats) uses the PC platform and the PC market is not willing to pay much for integration or framew^ork technology. Yet, as the company moves to compete more directly with "high-end" EDA vendors, its business model will likely demand higher average selling prices. Finally, DEC's major challenge will be to demonstrate added value beyond its framework.

The company is clearly heading in the right direction given the growth of its third-party software support activities (for example, its

Synergy program).

DEC is not the only non-EDA vendor entering the framework battle. Sun Microsystems Inc. also is developing framework technology, •which it has offered to the CFI for possible inclusion in the CFI's yet-to-be-released standards specifications. Semiconductor vendors, such as NEC

Electronics Inc., are bringing out their own frameworks. NEC recently demonstrated its

OpenCAD Design System, which incorporates

DEC'S Powerframe design environment, at the

Design Automation Conference.

Dataquest Perspective

Finding a Winning Framework

Solution

The big question that the market continues to ask is which framework will win? Dataquest believes that dear winners will not emerge for another three to five years.

Dataquest believes that the winners in the framework "wars" will not necessarily have the best framework, but rather will be able to

deliver the best EDA solution to the customer.

In this context, solution includes application software, component libraries, a range of services, support, and ftiture migration path as well as, of course, framework technology.

Most important, we also do not believe that ft^amework suppliers must create or own all of the value comprising the solution. This implies that framework suppliers should begin focusing more on delivering value and less on creating and/or owning value. This approach demands a fundamental restructuring of EDA companies' business models. The market is not concerned with where the value is created, so long as value is delivered in an uninterrupted fashion.

Thus, large EDA and framework vendors should begin to accept the idea of distributing certain application software packages from third parties and phasing out internally developed products that caimot remain competitive. EDA vendors should not invest an utueasonable amount of resources in developing application software, but instead concentrate on developing only those core applications that provide them with a strategic advantage such as account control.

The EDA companies that come up with the right mixture of added value will be able to deliver the best solution, and this will provide tremendous leverage to their frameworks. This, in turn, will translate to better account control and more stable revenue streams.

In any event, the whole notion of a shortterm winner or loser in the framework wars is without basis. •

By Ron Collett

Product Analysis

Silicon Graphics

Announces Indigo, a New Low-End

Workstation for Less than $10,000

(Silicon Graphics' 33-mips Indigo workstation lowered the price tag of 3-D modeling, offering implications for CAD/CAM/CAE users on workstation platforms. In gaining immediate support from two EDA vendors, SGI's Indigo machine also staked out new claims in the 2-D graphics market for Silicon Graphics, which historically had concentrated on higher-end 3-D applications. The following article is reprinted tvith permission of Dataquest' s Technical Computer

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA. 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

i

CAD/CAM/CAE

Systems group. In addition, there is an analysis by Dataquest's Michael J. Seely on how these recent events affect the CAD/CAM/CAE industry.)

Introduction

On July 22, Silicon Graphics Inc. (SGI) announced a suite of new products. The new products include IRIS Indigo, a new low-end workstation; IRIS Explorer, a visual application environment; CASEVision, a visual development environment; and the IRISserver family of file servers. This newsletter provides a synopsis of each announcement and Dataquest's view of the impact of the new Indigo workstation on the industry.

Announcement

IRIS Indigo

The major product that was announced on

July 22 was the IRIS Indigo. The system is based on a 33-MHz R30(X)A/R3010A, supports

8 to 96MB of memory, has up to 1.3GB of disk space, and has integrated digital audio tape

(DAT) quality audio capabilities. The system is rated at 26 SPECmarks and 4.2 mflops. Graphics performance was quoted at 250,000 2-D vectors/sec, 200,000 3-D vectors/sec, 14,000

Z-buffered flat-shaded triangles/sec, and 5,000 independent quads/sec.

A base system with 8MB of memory and a l6-inch color monitor lists for $7,995- With a

236MB formatted disk drive, the system is priced at $9,995. For the multimedia market, the

$26,000

Source: Dataquest (August 1991)

$105,000 typical base configuration will be the 16MB,

16-inch color monitor, and 432MB disk system, which lists for $12,500. The system will be available in volume in September.

SGI's Strategy is to expand into new application areas tvithin the technical marketplace.

An optional live video board is also available for the system. The expected availability date for the video board is the second quarter of

1992, and the board is expected to list for approximately $2,000.

IRIS Explorer

IRIS Explorer is a visual application environment that provides the end user with a means to view the numerical results of an engineering or scientific problem. The end user can simply connect various modules together in a pointand-click manner to apply various algorithms and graphics techniques to the problem without doing any programming. The package will initially include more than 100 modules. It will be available for current models and will become standard with all future systems using IRIX 4.0.

The delivery date is set for late 1991.

CASEVision

Table 1

IRISserver Family Base Configurations

Data Station 2 POWERfile 50

35-MHz R3000

16MB memory

780MB system disk

780MB expansion disk

Network backup software

Ethernet

IRK 4.0/NFS

Two 33 MHz R3000

32MB memory

780MB system disk

Three 1.6GB expansion disk

Network backup software

Ethernet

IRK 4.0/NFS

2.3GB Smm tape drive

1/4-inch tape drive

POWERflle 100

Four 33-MHz R3000

64MB memory

780MB system disk

Seven 1.6GB expansion disk

Network backup software

Dual Ethernet

IRIS 4.0/NFS

2.3GB 8mm tape drive

1/4-inch tape drive

Dual VME

(Juad SCSI channels

$165,000

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE

offered by SGI under this program. It is an integrated toolset for individual developers that includes a static analyzer, a debugger, and a performance analyzer. The CodeVision toolset will be available in 90 days, priced at $1,500.

IRISservers

SGI announced a new line of file servers: Data

Station 2, POWERfile 50, and POWERfile 100.

The systems' performance and basic characteristics are equivalent to the 4D/35, 4D/320, and

4D/340 servers but have had additional hardware, software, and networking features added to address the file server market. The base system and price for each model are shown in

Table 1. Hardware features have been added to the system, including dual POWER Channel architecture, a SCSI expansion cabinet, a 1.6GB

SCSI disk, a 1/2-inch SCSI tape drive, and a dual channel VME/SCSI controller. The new software includes network backup software, volume management software, mirroring software, network archiving software, and IRIX 4.0. All systems are currently available.

Dataquest Perspective

The most important aspect of this whole announcement is that SGI finally has a box for under $10,000. Sun Microsystems Inc. has aggressively pursued and cornered j h e low-end desktop market. Therefore, in order to compete for the desktop, a vendor has to have a box for under $10,000. Now, SGI can compete in this sector of the market. The only other leading competitors offering RISC workstations in this price range are Sun and Digital Equipment

Corporation. However, we expea new products to be announced by Hewlett-Packard Company and IBM Corporation at this price point before the end of the year. Thus, SGI has litde time to capitalize on this new product.

The second important feature is that this system is a competitive 2-D box—a first for SGI. With the majority of applications still only requiring

2-D capabilities, end users had to pay a premium for the 3-D graphics. As such, SGI was only able to penetrate accounts requiring

3-D graphics. The new system, which is competitive both in terms of performance and price with other 2-D systems available today, will broaden the total available market for SGI. And, it also puts SGI in a better position for upgrade business because it is always easier to upgrade a user to 3-D, video, or audio capabilities than to have to oust the incumbent. But more importandy, the jKJtential for increased volumes combined with the Compaq Computer Corporation and Microsoft Corporation deals should help SGI attract new software vendors to its platform.

Additionally, SGI has an advantage over every major workstation vendor, except Digital, with its symmetric multiprocessing servers. The introduction of a server line that takes advantage of the multiprocessing as well as provides the improved functionality, such as disk backup and disk management, is a very good move on

SGI's part. According to Dataquest's Server group, file servers in 1990 represented the largest usage of servers in terms of dollars. File servers and database servers are projerted to be the two largest segments by 1995.

The major drawback to this announcement was

SGI's emphasis on the audio and video capabilities of Indigo. Although these products offer the hardware functionality, no software is currently available that takes advantage of all these features—not even e-mail. Because of the lack of software and the faa that the multimedia market is still in its infancy, we do not e x p e a these features to stimulate a huge increase in demand for SGI's products in the near future.

However, •what audio and video capabilities will do is position the company at the forefront of multimedia technology in preparation for growth in this market.

A second area that was not addressed was the distribution channels for selling these systems.

With the lower price point, distribution through the direct channels becomes quite expensive. At this time SGI is not in a position, based on its unit volume, where it could entice other channels to offer its systems. However, it must still address the issue of distribution channels or suffer the consequences of lower profits.

In conclusion, SGI's strategy is to expand into new application areas within the technical marketplace. The price point and 2-D capabilities of Indigo wiU increase SGI's total available market. However, with the multimedia capabilities,

SGI may find itself actually serving more corporate commercial applications rather than technical ones because die most promising areas for multimedia are in corporate training and presentations.

From the beginning, Silicon Graphics has positioned itself as a 3-D graphics supplier, staking its claim on the high road of graphics workstations. With the introduction of Indigo, SGI has effectively brought the entry point of 2-D and

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE

3-D graphics down to a level where sophisticated 3-D modeling may begin to reach the masses.

In addition, SGI immediately garnered endorsements from the CAD/CAM/CAE market with electronic design automation (EDA) vendors

Cadence Design Systems Inc. and Valid Logic

Systems Inc. announcing ports to the SGI machine. With the Cadence and Valid ports, SGI can now go after opportunities in the EDA market that historically have been 2-D-only applications, such as printed circuit board (PCB) layout. Designers of electronics produas can use the Indigo and higher-end machines to operate sophisticated 2-D and 3-D electronic package design software. The ports also will eventually open up the prospect of future 3-D modeling for a variety of elearonic design problems. In the long term, even chip design may become a 3-D application with detailed design and analysis at the micron level. •

By Laura Segervall

Michael J. Seely

Highland Software

Ships Software Store

(With Highland Software's shipment of the Software Store, a collection of UNIX application software will now be available to users of

UNIX-based workstations in the CAD/CAM/CAE area. Representing a promising distribution method for CAD/CAM/CAE software, the Highland Software program has the potential to benefit both software vendors and users alike in the CAD arena. The following article, written by an analyst in Dataquest's Personal Computer

Software group, is reprinted with permission.

Additional analysis is provided by Dataquest's

Kathryn Hale on how these recent events affect the CAD/CAM/CAE industry.)

The first volume of the Software Store, a CD-

ROM containing UNIX application software packages and text and graphics exhibits from many independent software vendors (ISVs), shipped this summer from Highland Software.

Volume I of the Software Store CD contains more than 80 products from 36 companies.

Software areas covered include CAD/CAM/CAE, office automation, electronic publishing, graphics and visualization, and system utilities.

There are three versions of the Software Store— the Software Store for ULTRDC, the Software

Store for HP, and the Software Store for Sun.

Additional volumes are to be added quarterly.

Each edition of the Software Store packages 20 to 35 applications on a single CD for a single hard-ware platform. The editions are published quarterly.

CD-ROM technology offers CAD vendors significant opportunities to reach and influence prospective users electronically.

ISVs purchase space by the megabyte and a start-up kit that includes the following:

• Software utilities

• Documentation

• Special support for first-time exhibitors

• Beta disk for quality assurance

The Distribution Method

The CDs are distributed at no cost to qualified users of workstations based on the UNIX operating system, to computer company sales offices, and to computer distributors. The prospect inserts the CDs in the CD-ROM drive to examine, evaluate and, in some cases, to immediately purchase the software.

For the ISVs, the Software Store serves as a worldwide marketing, distribution, and promotion medium. It carries both demonstration and licensable, password-protected UNIX software.

The customer contacts the ISV direcdy to discuss purchase, removing the burden of the middleman from Highland Software. Each company informs users of the contact name for produa and purchasing information.

Easy t o Use

Users insert the CDs into CD-ROM drives connected to workstations and "mount" the CD-

ROM as part of the computer's file system.

Some programs may be purchased or evaluated by contacting the ISV for a password. Installation is automatic. Each CD contains an electronic catalog of UNIX application software with password protection, demonstration versions of software, and on-line documentation. ISVs maintain control of the presentation, promotion, and distribution of the software exhibits. Highland provides a toolkit, technical support, standard software installation procedures, tools to help create exhibits, and integration of the exhibits on the CD.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE

8

CD-ROMs—The Preferred Method o f Distribution?

Other companies using this method of promoting and selling their products include Digital

Equipment Corporation for its RISC/ULTRIX products, Hewlett-Packard Company for its HP

9000 family of product, and Sun Microsystems

Inc. for its SPARC computers.

$400-$6D0 r-^c^

SJ^ v>>C (300-S400

V\?

K ^

^

I

$100-$150

Customer Sales

Office Office

'^& sso-sioo

Source: Highland Software

^

Manuals

$1-$10

Trade Demo Software

Show Tape/ Store

The Software Store incorporates the X Window graphical user interface, allowing users to pointand-click on product demonstrations. Following the subsequent purchase and unlocking of the software through the use of password codes provided by the software vendor, users also have the ability to personally install the software. The CD-ROM also includes text-retrieval software, which allows users to locate software by typing in key words that suggest particular applications.

An obvious qualifying tool, CD-

ROM distribution could replace both a portion of advertising in other media as well as a large number of initial sales calls.

Although CD-ROM distribution will reduce the cost of sales by automating portions of the job of selling, it also will make selling CAD software much more competitive.

Figure 1 shows the approximate costs to an ISV to promote its product through the usual methods. These costs are incurred before a prospea is qualified or any sale is made. The comparison points out very vividly the savings when using this type of method to make that first important contaa.

Selling software directly is the most expensive way of getting a produa to the prospect. Other methods are less costly but continue to keep the price of the software high and limit sales.

FIGURE 1

Costs per Customer Demonstration

To exhibit on one platform on the Software

Store, chaiges would be $5,360 for a 3MB demonstration and $15,000 for a 25MB program, in addition to a $3,000 one-time charge. Shelf space on additional computer brands is available at reduced rates, and additional discounts are available now.

Dataquest Perspective

CD-ROM technology offers CAD vendors significant opportunities to reach and influence prospective users electronically. An obvious qualifying tool, CD-ROM distribution could replace both a portion of advertising in other media as well as a large number of initial sales calls. The technology also gives the converted prospea ready access to an "internal selling" tool. We foresee an increase in political skirmishes within a prospective buyer's ranks, as internal champions of competing solutions are able to demonstrate, at will, the good and bad points of various products.

Some potential also exists to use CD-ROM delivery both to attract new dealers and to integrate resellers better with a vendor's direa sales force. With effective, interaaive CD-ROM produ a demonstrations, dealers could more readily step up to front-line selling of sophisticated workstation-based software, bringing in the vendor's direa sales staff for highly qualified prospects. This scenario would require, of course, creativity in compensation structures.

Although CD-ROM distribution will reduce the cost of sales by automating portions of the job

©1991 Dataquest Incorporated / 1290 Ridder Parle Drive, San Jose, CA 95131-2398 / (408) 437-6000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE of selling, it also will make selling CAD software much more competitive. The technology invites a "free trial," potentially exposing products to increased testing. We have long advised CAD vendors to emphasize creating products that are easy to learn, easy to use, and bug-free. We have also stressed the importance of building service, support, and training into the product itself, rather than relying on individual performers alone to supply these key services. Those vendors with products that best meet these requirements are now in a superior position to leverage CD-ROM distribution, an environment where the product that stands out is the product that performs best during a prospective buyer's uncontrolled, untutored exploration. •

By Gladys Francis

Kathryn Hale

three contingents favoring either the Apple/

Macintosh, IBM/Intel, or SPARC/RISC/Motorola architectures for heavy-duty document imaging.

DIMS vendors, integrators, and software developers tend to support one particular workstation/server architecture for a number of diverse reasons, including performance, display capabilities, networking, multitasking, distributed computing, of>erating system, adherence to industry standards, price, and the preferences of their customers.

Given the fact that Sun workstations were developed with sufficient horsepower to handle scientific and technical applications, document imaging is a very low end application in the Sun universe.

Company Analysis

Workstation Vendor

Strategies in the

DIMS Market-

Sun Microsystems

We hear a lot these days about the "Woricstation Wars," technological skirmishes wherein the various workstation vendors vie for supremacy on the desktop. An ongoing battle pits Intel

PCs against SPARC/RISC/Motorola technical workstations. Each camp makes claims about superior performance, compatibility, networking, and market penetration. The truth is that the distinctions between PC and technical workstation are blurring. Multitasking operating systems for the PC such as OS/2, and graphical user interfaces such as Windows 3-0 and the Presentation Manager, make PCs look and behave ever more like workstations.

But more is at stake in these wars than the desktop. Networked, distributed computing is made possible by high-performance servers that take the place of minicomputers and even mainframes. Workstation and PC vendors are now in the server business, and distributed computing has become big business.

Document image management systems (DIMS) providers cheer these battles on from the sidelines. The DIMS industry is divided roughly into

This is the first in a series of articles exploring the strategies of workstation/server vendors to position their products in the DIMS market. Our first workstation warrior is Sun Microsystems

Inc. of Mountain View, California.

Sun's Philosophy—Keep It

Simple, Keep It Separate

Sun is first and foremost a manufacturer—of technical workstations, servers, and the recently introduced SPARCprinter. To assure market acceptance of these products and ensure that they operate in the most efficient manner possible, Sun's secondary business is developing operating system software, networking software, user interfaces, and application development environments compatible across all Sun products. All Sun products are based around a single microprocessor architecture—SPARC—thus simplifying development and support.

Sun's customers are systems integrators, large end users, and independent software vendors

(ISVs). All add value to Sun hardware by developing software to run on it, by building systems around it, or both. To facilitate this development. Sun provides its partners with tools including a high-level programming language, development tools, and communications software.

Sun's marketing strategy aims to position Sun products as the most high-performance, compatible, flexible, and cost-efficient available. Sun's support organization is chartered to provide customers with all pre- and postsale support

©1991 Dataquest Incoiporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

10

CAD/CAM/CAE needed to sell and install an account successfully.

Overall, we believe that document imaging, as a horizontal application, ivill be tightly integrated into desktop computing environments over the course of the 1990s.

A recent restructuring at Sun spun off two new companies chartered with developing and managing the company's investments in UNIX software development. This diversification created separate operating entities responsible for managing software (rather than hardware) development activities, as follows:

• SunSoft Inc. is chartered to manage Sun software development activities including the Sun operating system (SunOS), windowing system

(Xll/NeWS and the OpenWindows environment), and networking software supporting the Network File System (NFS) and Open

Network Computing (ONC) architecture. The group wdU concentrate on sales, marketing, and licensing of these technologies and work with international standards bodies to ensure that Sun software produas comply with industry standards.

• SunTech Enterprises Inc. supports third-party developers with tools and products to build applications and systems around Sun hardware. Products managed by SunTech include

SPARCprinter, NeWSprint software, PC Network File System, SPARCompiler software development platform, and SunLink connectivity gateways to systems and networks.

Sun's Imaging Strategy

As the market for commercial and technical document imaging has heated up, so has Sun's interest in positioning its workstations there.

Sun has targeted five main areas of opportunity in document image management, as follows:

• Document imaging—^The scanning, storage, retrieval, and manipulation of business-size paper documents

• Engineering data management—The scanning, storage, retrieval, and manipulation of engineering drawings and related documents such as bills of material and design specifications

• Full-text search—^Locating information using software that searches for words and phrases specified by the user

• Facsimile—Sending and receiving faxes through a workstation

• Multimedia/compound documents—Integrating a variety of media types including text, data, graphics, scanned images, video images, and voice into one on-line "document"

Sun's Imaging Product Line

Sun's main product line consists of a family of workstations and servers built around the

SPARC microprocessor architecture, running under the UNIX operating system, and adhering to the client/server computing model.

Sun appears to be in the right place at the right time with a wellengineered solution for bringing general-purpose document imaging to the UNIX desktop.

Sun's original products were high-end technical workstations designed for use in scientific and engineering applications. To take advantage of emerging applications for workstations in the commercial market, over the past 18 months

Sun has introduced a series of lower-cost, desktop workstations aimed at business customers.

Base prices for Sun's high-end workstations, typically used for technical applications, range between $29,900 and $69,900. By contrast, low^er-end workstations for the commercial market cost much less. The diskless SPARCstation

SLC sells for a base price of $4,995, and the

SPARCstation IPC sells for $7,995- Base prices for the four servers in the SPARCserver family range from $24,590 for the SPARCserver 2 to

$99,900 for the SPARCserver 490.

In September 1990, Sun introduced the SPARCprinter, a 12-pages-per-minute (ppm) desktop laser printer powered by Sun NeWSprint software. Newsprint builds on Sun's Network

Extensible Window System and offers PostScript compatibility. Printing has always been something of a bottleneck on Sun's otherwise highperformance systems. By solving this through a software solution, Sun users can take advantage of the high throughput potential of the SPARC architecture. The printer and software are bundled for a list price of $2,695-

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE 11

The company believes that its SPARCstations and servers are ideal platforms for handling document imaging applications, dting the following reasons:

• Performance—^Technical workstations generally offer high-performance capabilities required to run image-processing applications, such as: fast processing sp>eeds; extended memory; multitasking; ample disk storage; highresolution, large-screen displays; and built-in networking capabilities.

• Networking—Server-based document imaging networks can be set up to grow as the user's needs grow, to support other applications in addition to document imaging, and to tie into other computer systems and databases within the corporation. In addition,

Sun's ONC architecture ties computers from other vendors into the Sun environment.

• Distributed computing—^The client/server model provides an efficient style of computing in DIMS environments—^if one server goes down, other functions can proceed uninterrupted.

• Database management—^High-performance servers are well suited to handle image database management functions in installations comprising lai;ge numbers of image-intensive workstations.

• Multitasking—Sun workstations combined with the UNIX operating system enable users to perform several tasks concurrendy on a workstation—^for example, to simultaneously view an image, pull in data from a mainframe database, and integrate this information into a word processing document.

• Multimedia—Sun workstations are designed to display and integrate mixed data types including text, data, graphics, scanned images, video images, and voice.

• Price—Costs are declining to the point where

Sun workstations are competitive with Intelbased PCs often used in image-processing environments (see Table 1),

Sun's Imaging Partners

Sun relies on systems integrators and software developer partners to integrate its workstations and servers into document imaging installations.

The integrators usually work as prime contractors, pulling Sun into the bid if the customer's installation warrants a Sun-type solution. In the case of software developers that port software to Sun platforms, sales may be driven by the software partner, an integrator using the software solution, or Sun's sales force.

Sun works with leading integrators to install commercial and technical DIMS. Sun's commercial DIMS integrators include Andersen Consulting, Electronic Data Systems (EDS), FileNet,

Table 1

SPARCstations versus PCs

Configuration

List/Street Price ($)

Mips

Monitor

Price

Monitor Controller/Driver Price

OS/Windows Price

Cable Price

Image Adapter Board Price

Network I/F Price

Mouse Price

Total

Standard

PC

200 MHz,

386/2MB,

RAM/60MB disk

I m a g i n g PC

25 MHz, 386/4MB,

RAM/60MB disk

SPARCstatlon

SLC

20 MHz,

SPARC/8MB,

RAM/diskless

1 P r o d u c t i o n

I m a g i n g PC

33 MHz, 386/8MB,

RAM/IOOMB disk

5,600/3,500

4.5

14" VGA, 57 dpi

6,800/3,500

6.2

4,995

12.5

6,500

8

14" Super VGA, 70 dpi

17" 1152 X 900,

100 dpi

19" high resolution, 115 dpi

SPARCstation IPC

25 MHz, SPARC/

8MB, RAM/

207MB disk

7,995

15.8

19" high resolution, 115 dpi

Included

Included

Included

Included

NA

$450

$100

$4,100

$450

Included

$270

$155

$950

$450

$100

$5,875

Included

Included

Included

Included

NA

Included

Included

$4,995

$2,050

$1,900

$270

$155

$1,500

$450

$100

$12,925

$2,700

Included

Included

Included

NA

Included

Included

$10,695

NA = Not available

Note: Street prices are generally unavailable on production imaging PCs because of the need to work with an integrator.

Source: Sun Microsystems Inc., Dataquest (August 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

12 CAD/CAM/CAE

Genesis Imaging Technologies, Grumman

InfoConversion, NYNEX, Perot Systems, Philips

Information Systems, Price Waterhouse, TRW

Financial Systems, and Xerox. Sun's technical

DIMS integrators include Cimage, FORMTEK,

GTX, Litton/Integrated Automation, Optigraphics, and Planning Research Corporation (PRC). Table 2 lists leading software partners that have ported document imaging solutions to Sun platforms.

(Sun's partners place a n order for a certain number of workstations and servers, but d o not necessarily tell Sun w h e r e they are going). Sun recendy listed 77 d o c u m e n t imaging installations in place worldwide using Sun workstations or server/workstation combinations. Approximately

62 of these systems are installed in the United

States, 13 in Europe, a n d 2 in t h e Far East.

Imaging Installations

Though the exact n u m b e r of imaging installations m n n i n g Sun hardware is difficult to track

Table 2

S u n ' s I m a g i n g Software Partners

Partner

Calera Recognition Systems

JTS Computer Systems

NYNEX Image Recognition Systems

OCR Systems

Plexus

Resumix

Workgroup Technologies

Xerox Imaging Systems

Source: Sun Microsystems Inc.

As s h o w n in Figure 1, t h e majority of Sun's reported document imaging sites are in the manufacturing sector. Users in this industry include Boeing, Ford, GE Aircraft Engines, Mack

Product Ported t o Sun Environment

OCR software and developer's kit

OPENIMAGE document management solution for paper-intensive environments

Intelligent character recognition software that reads handprinted characters

ReadRight OCR software

Plexus XDP image-processing software on Sun platforms

Resumix resume-processing and pyersonnel management software

Technical data and document management software

ScanWorX text recognition and image capture system

Figure 1

Sun's D o c u m e n t I m a g i n g Installed Base b y Industry

Telecommunications

Utilities 10%

Pharmaceutical/

Health Care 8%

Legal 6%

Financial Services 4%

i

I

Source: Sun Microsystems Inc.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE

13

Truck, Martin Marietta, Pratt & Whitney, Rockwell Space Division, and Rolls Royce. Government sites comprise 18 percent of installations, with major U.S. federal customers including the

Department of Defense, the Environmental Protection Agency, the Library of Congress, the

U.S. Navy, and the U.S. Secret Service; and dty government customers including the City of

Chicago's Bureau of Parking and the New York

City Board of Elections. Insurance and banking account for 12 and 11 percent of Sun installations, respectively. Major bank customers include Bank of America and Citicorp Mortgage.

Insurance providers include Blue Cross/Blue

Shield, Great Western Life, and New York Life.

Major customers from other industries include:

Merrill Lynch and Fidelity Investments (financial services); the University of California, Berkeley

(education); J.C. Penney, Northwest Airlines, and

Walt Disney Pictures (general business); Bristol-

Meyers Squibb and Merck, Sharp and Dohme

(pharmaceuticals); and AT&T, British Telecom, and Pacific Bell (telecommunication).

The Bottom Line

Document imaging is a steadily growing business at Sun. In Sun's 1990 fiscal year, sales of workstations and servers into document imaging installations accounted for $100 million of Sun's total revenue. In fiscal 1991, this number had grown to $140 million, an increase of 40 percent. This represented 5 percent of Sun's total revenue for fiscal 1991 (roughly 3 percent was commercial DIMS and 2 percent was technical

DIMS).

Dataquest Perspective

Given the faa that Sun 'workstations were developed with sufficient horsepower to handle scientific and technical applications, document imaging is a very low end application in the

Sun universe. If anything, Sun workstations are probably "overqualified for the job." But their fast {performance, multitasking capabilities, and integrated networking, combined with increasingly attractive list prices, make them evermore popular choices for managing DIMS databases.

Judging from the distinguished list of partners that work with Sun to provide document imaging solutions, the Sun workstation/server model is becoming an industry standard among

UNIX-based DIMS providers.

We believe that Sun will successfully further penetrate this market for the following reasons:

• Consistent quality of product

• Ever-increasing price/performance ratios

• Dependable OS/networking software services to support the Sun client/server model

• Ability of Sun systems to reside peacefully and share information in multivendor environments

• Sun companies chartered to develop and support third parties and ISVs

• Support organizations within Sun focused on customer service

• Consistent focus from the corporate level down toward development of a single, integrated product line

Overall, we believe that document imaging, as a horizontal application, will be tighdy integrated into desktop computing environments over the course of the 1990s. Systems offering superior networking and the ability to handle a variety of compound data types will lead this movement. Sun may well be at the vanguard of this trend. Its competitors face unresolved problems: The Intel PC crowd dithers over the acceptance of OS/2; proprietary systems vendors struggle to open up dosed architectures, support mukiple platforms, and move to UNIX at the same time; the Madntosh provides applications consistency and gorgeous graphics capabilities, but so far is still a relatively proprietary solution and has achieved low penetration of the general business and technical markets. Sun appears to be in the right place at the right time with a well-engineered solution for bringing general-purpose document imaging to the

UNIX desktop.

One note of caution: The recent restructure that spun off the SunSoft and SunTech companies has left gaps to be filled. Many employees find themselves not only working in new jobs, but at new companies as well. In some instances, replacements have not been identified or may not be hired, leaving gaps in the previous infrastructure. This restruaure affected the marketing team for Sun's commerdal document image systems development, for example, but left the TDIMS team relatively intact. Although the formation of SunSoft and SunTech will promote better long-term support for Sun customers, we expect it to take some months before business returns to normal. •

By Pamela Stone Bliss

(This article is reprinted with the permission of

Dataquest's Document Imaging service.)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

14

CAD/CAM/CAE

News and Views

Cadence and Siemens

Nixdorf Enter Joint

EDA Venture

Cadence Design Systems Inc., of San Jose,

California, and Siemens Nixdorf Informationsysteme AG (SNI), of Paderborn, Germany, last month announced their intention to combine operations in central Europe to jointly develop, market, and support EDA solutions for the

European market. The joint venture's headquarters will be in Munich, Germany, and will include the combined sales, marketing, and service oiganizations of the two companies.

The new company, called Cadence Europe

GmbH, will be jointly owned by Cadence and

Siemens Nixdorf, with Cadence holding a oneshare majority stake. The combined oi;ganization aims to leverage the technology and expertise of the companies in order to provide a set of products based on a unified framework,

Cadence's Design Framework II.

The Cadence Perspective

Dataquest believes that Cadence will accrue the following benefits from this joint venture:

• Through this deal, Cadence potentially gains greater access to Europe's biggest EDA market: Germany, which accounted for 33 percent of all EDA sales in Europe in 1990. In the longer term, entry into Eastern Europe will be eased due to Germany's (and Siemen's) special links with the East.

• Cadence positions itself as the champion of open framework standards. Its association with Siemens, a leader in the Jessi European framework standards effort, will enhance its image as a key player in the framework scene.

• Cadence has moved from a niche IC layout company to embrace CAE and printed circuit board (PCB) layout. SNI owns key technology—^particularly in the systems, CAM, testability, data management, and EMC areas.

If properly managed, this technology could enhance Cadence's product offerings considerably, moving the company further into the broad electronic product development field.

The Siemens Nixdorf Perspective

By joining forces with Cadence, SNI gains the following:

• Assistance in the marketing and support of products within the German region

• Accelerated returns on investment in EDA tools (The rate of returns following the Calay

Systems Inc. acquisition and substantial internal development was proving slow for SNI.)

• A ready-made, credible product line for sales into the Siemens organization (Siemens' increasing need for IC development tools is likely to have been a major factor in this respect.)

Dataquest Perspective

This deal follows Cadence's joint venture with

European Silicon Structures last year, when a

European EDA development and marketing center was established in the United Kingdom.

Cadence is dearly strengthening its underdeveloped distribution channels in Europe (in comparison with other broad-line EDA vendors) through a strategy of joint venture and direct channels to suit local market conditions.

The deal gives Cadence greater access to the

SNI organization, which is an appreciable slice of the German market. However, SNI does not operate a 'single-vendor' EDA jralicy, and

Cadence will have to fight alongside other suppliers—notably Racal-Redac Inc., Dazix Intergraph, Mentor Graphics Corporation, and Valid

Logic Systems Inc., all of which have sizable user bases within SNI. The deal is also a valuable springboard into the wider German EDA market, which was valued at $318 million in

1990, using Cadence's strengths in other geographical regions as an additional credibility factor.

From SNI's viewpoint, the following possibilities exist:

• SNI realized it could not survive in the EDA business in view of its late entry and, therefore, has effectively sold out to Cadence. We do not believe this to be true. More likely,

SNI is serious in its intentions to become a world EDA player. SNI usually takes a longterm strategic view, and the company would not have acquired the assets of Calay Systems last year if it was not serious in its intentions (although the future of Calay itself is now open to speculation).

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE 15

• SNI found itself in a vulnerable position because many of its products are sourced from competitors. The company did not have the funds to develop or acquire replacement produas, and, therefore, teamed up with a company that had a strong and urgent need to increase its share of the lucrative German market. This latter point places SNI into a strong bargaining position.

• SNI needed to extend its operations on a wide geographical base. The company decided to achieve this by tapping into the technology and culture of Cadence, one of the rising stars of the EDA world. In this scenario, one thing does not add up: If the venture covered a different geographical region (for example. North America or Asia),

SNI would be gaining access to markets that are essential if it wishes to become a world player in EDA. However, the agreement is restricted to the geographical region around

Germany—SNI's home territory. We believe that there must be more in this for Siemens

(such as access to Cadence's wider distribution channels) and that further announcements will follow.

In the short term, this agreement will have little effect on the European EDA market. In the longer term, we believe that Cadence's position in Europe wiU be enhanced by it. One of the most interesting questions relates to the culture of the two companies; Will the fast-moving, dynamic Cadence and the laige, slower-moving

SNI result in a happy marriage, or, instead, will there be a serious personality dash? •

By Jim Tully

©1991 Dataquest Incoiporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

16

In Future issues

Topics of interest to the CAD/CAM/CAE industry will appear in upcoming issues of Dataquest

Perspective, including the following:

• A report on the mechanical applications market

• An overview of the European electronic design automation (EDA) market

More "News and Views" on the CAD/CAM/

CAE industry

CAD/CAM/CAE

i i

For More Information . . .

Managing Editor Patrick Waurzyniak (408) 437-8132

On the topics in this issue CAD/CAM/CAE (408) 437-8132

On related subjects Client Inquiry Center (408) 437-8671

About other Dataquest publications Sales (408) 437-8000

About upcoming Dataquest conferences Conferences (408) 437-8245

About your subscription Customer Service (408) 437-8402

Via fax request Fax (408) 437-0292

The content of this report represents our Interpretation and analysis of Information generally available to the public or released by responsible Individuals In die subject con^tanlea, but is not guaranteed as to accuracy or con^leteness. It does not contain material provided to us in confidence by our clients. Individual companies reported on and analyzed by

Dataquest may be clients of this and/ctf other Dataquest services. This information is not furnished in connection with a sale or offer to sell securities or in connection with the solicitation of an c^er to buy securities. This firm and its parent and/or their officers, stockholders, or members of their families may, fitom time to time, have a long or ahoit position in the securities mentioned and may sell or buy such securities.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

i

Dataoyest

3 company of

TtK Dun 8L Bndstrcct CorporatH

Dataquest

Perspective

CAD/CAM/CAE

Vol. 1, No. 4

August 26, 1991

Marta't An.ilysis

Impending Framework Fragmentation Triggers New EDA Business Strategies

Contrary to electronic design automation (EDA) users' expectations, a single, homogenous, and open framework for the software tools used in product design is not likely to happen soon, if at all. The following Dataquest research examines the impending fragmentation of the EDA framework market and its impact on the business strategies of EDA vendors.

By Ron Collett Page 2

V\u(.h\c\ Aiiiilvsfs

Silicon Graphics Announces Indigo, a New Low-End Workstation for Less than $10,000

With the latest workstation from Silicon Graphics Inc. (SGI), Indigo, SGI effectively brought the entry point of 2-D and 3-D graphics down to a level where sophisticated 3-D modeling may begin to reach the masses. In addition, the SGI workstation immediately gained support from the CAD/CAM/CAE market with EDA vendors Cadence Design Systems and

Valid Logic Systems announcing ports to the machine, opening up the prospect of future

3-D modeling of chip design for the EDA market.

By Laura Segervall and Michael J. Seely Page 4

Highland Software Ships Software Store

Highland Software is shipping the Software Store, a compendium of UNIX application software on a CD-ROM disc that is distributed at no cost to qualified UNIX workstation users for evaluation. Offering independent software vendors the chance to demonstrate their

UNIX software to prospective buyers, the Highland Software program currently distributes three versions aimed at users of UNIX software on workstations from the Hewlett-Packard

Company, Sun Microsystems Inc., and Digital Equipment (Corporation.

By Gladys Francis and Kathryn Hale Page 7

C-OMip.iiiv Aotitvsis

Workstation Vendor Strategies in the DIMS MarketSun Microsystems

Sun Microsystems Inc. reported document imaging revenue totaling $140 million for its 1991 fiscal year, a 40 percent growth over last year. The company is making successful inroads into the document image management systems (DIMS) market through persistent adherence to standards in product development, marketing, customer support, and channel cultivation.

This article examines Sun's imaging strategy and comments on the company's chances for long-term success in the DIMS market.

By Pamela Stone Bliss

Page 9

News <intl Views

An analysis of recent news events affeaing the CAD/CAM/CAE industry

Cadence and Siemens Nixdotf Enter Joint EDA Venture

Page 14

©1991 Datacpiest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE

Market Analysis

Impending Framework

Fragmentation Triggers

New EDA Business

Strategies

Dataquest's research indicates that the EDA market has been somewhat preconditioned to believe that a single, homogenous, open framew^ork for electronic design will be established in the short term. The market anticipates that this framework will emerge from the myriad of vendors marketing frameworks and the CAD

Framework Initiative (CFI).

Dataquest, however, does not expect this to occur for at least five years, if at all. Rather, we believe that the framework market will undeigo considerable fragmentation over the next two to three years. In short, a stampede of different suppliers' frameworks will penetrate the EDA user base, leaving a fragmented market in its wake (see Figure 1). Although frameworks from different suppliers will have a different

"look-and-feel," most, if not all, will incorporate the standards ratified by the CFI. As a result, the major incompatibilities among frameworks that will emerge in the short term will diminish over time. The foUow^ing research examines the dynamics of this splintered framework market and offers a perspective on what it means for

EDA suppliers in the future.

EDA Frameworks: A Lack of

Standards

In spite of the EDA market's need for standards, a salient fact is that few standards exist today. This is particularly true in the area of

EDA frameworks. A framework comprises a multiplicity of capabilities, but Dataquest's research indicates that both the industry and the market have varying views on exacdy what constitutes a framework. However, staying within the guidelines pronounced by the CAD

Framework Initiative, a framework comprises the following eight elements (for which CFI has formed technical committees):

• Architecture—^The interdependencies and interrelationships among functional areas of a framework

• System environment—Operating system services necessary to provide hardware independence to other subsystems within a framework

Figure 1

Framework Fragmentation

CAD Framework Initiative

The EDA

Frameworl< Wars and Standards

CAD Frameworl^ Initiative

1991 1992

Source: Dataquest (August 1991)

1993 1994 1995

1996

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE

Design data management—The mechanisms for storing, accessing, and versioning of design data

Design methodology management—^The set of activities necessary to create and complete a design

Design representation—^The conceptual data modek used to describe the various elements of a design

Intertool communication—^The mechanisms for efficient sharing of design information between tools

User interface—^The portion of the system that manages all interactions with the designer or other user

Component information representation—^The digital representation of electrical components including models, libraries, and distribution of component information

The big question that the market continues to ask is which framework will win? Dataquest believes that clear winners will not emerge for another three to five years.

There has been a proliferation of framework announcements from many different vendors, which is leading to a more intensified and protracted battle for framework market share.

Among the most aggressive vendors digging in for the impending framework wars are Mentor

Graphics Corporation, Cadence Design Systems

Inc., Valid Logic Systems Inc., Dazix/Intergraph,

Racal-Redac Inc., Viewlogic Systems Inc., and

Digital Equipment Corporation (DEC).

What does supplying the framework to the customer mean strategically for EDA vendors? Dataquest believes that it means account control— the vendor that supplies the framework will essentially "own" that customer. Account control more or less equates to an ongoing revenue stream from that particular customer. We anticipate that customers will look first to their framework supplier when investigating the purchase of additional products and services. Thus, for lai^e EDA vendors whose businesses demand a high revenue stream, the framework is of enormous strategic importance. But the marketplace, we believe, will continue to exhibit near indifference to firamework offerings in the absence of EDA application software that is of superior capabilities and integrated into the framework. In other words, the framework alone is of little value without a robust suite of tools. Yet the framework itself will have to stand on its own merit. Therefore, EDA vendors must deliver both in order to establish a framework beachhead.

Dataquest believes that the winners in the framework "wars" will not necessarily have the best framework, but rather will be able to deliver the best EDA solution to the customer.

Dataquest expects the North American EDA framework market to be 60 to 70 percent penetrated by 1995, with Asia and Europe following by 1996. In our view, EDA vendors currently possessing the lai^est installed base will wield both a tactical and strategic advantage over competitors in the imjjending framework battles. Of course, it is important to evaluate installed base within the context of account control—the value of a laige installed base is seriously diminished if account control is lacking within the base.

Using installed base as a proxy for competitive advantage. Mentor is at the top with a base of approximately 22,000 seats, following by Valid with approximately 15,000 seats, and Dazix/

Intei;graph with about 11,(X)0 seats. Other contenders include Cadence, Viewlogic, and

Racal-Redac.

The market is not concerned with where the value is created, so long as value is delivered in an uninterrupted fashion.

In spite of Cadence's sheer size and the fact that it has high brand recognition in the framework arena, the only market where the company wields a significant edge is the IC layout sector. Here the company has approximately

4,000 seats, which correspond to approximately

36 f)ercent market share. Yet in the systems market. Cadence's strength is tenuous. The company's presence has been limited almost exclusively to the Verilog simulator, and the Verilog simulator has traditionally resided with another

EDA vendor's environment. Thus, at best,

(Cadence shares account ownership of its Verilog

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE installed base with companies such as Mentor and Valid. We believe that this puts Cadence at a significant disadvantage in the framework market share wars, although it is reasonable to assume that some of Cadence's power in the

IC layout market will spill over into the systems arena.

Racal-Redac is also at a disadvantage in that it, too, shares ownership of its customers with many other suppliers. Its installed base is split into two distinct camps—the HHB Systems CAE installed base and the Racal-Redac printed circuit board (PCB) layout base. Viewlogic faces similar challenges in the framework wars because the majority of its installed base (approximately 8,000 seats) uses the PC platform and the PC market is not willing to pay much for integration or framework technology. Yet, as the company moves to comjjete more directly with "high-end" EDA vendors, its business model will likely demand higher average selling prices. Finally, DEC's major challenge will be to demonstrate added value beyond its framework.

The company is clearly heading in the right direction given the growth of its third-party software support activities (for example, its

Syneigy program).

DEC is not the only non-EDA vendor entering the framework battle. Sun Microsystems Inc. also is developing framework technology, which it has offered to the CFI for possible inclusion in the CFI's yet-to-be-released standards specifications. Semiconductor vendors, such as NEC

Electronics Inc., are bringing out their own frameworks. NEC recently demonstrated its

OpenCAD Design System, which incorporates

DEC'S Powerframe design environment, at the

Design Automation Conference.

Dataquest Perspective

Finding a Winning Framework

Solution

The big question that the market continues to ask is which framework will win? Dataquest believes that clear winners will not emerge for another three to five years.

Dataquest believes that the winners in the framework "wars" will not necessarily have the best framework, but rather will be able to

deliver the best EDA solution to the customer.

In this context, solution includes application software, component libraries, a range of services, support, and future migration path as well as, of course, framework technology.

Most important, we also do not believe that framework suppliers must create or own all of the value comprising the solution. This implies that framework suppliers should begin focusing more on delivering value and less on creating and/or owning value. This approach demands a fundamental restructuring of EDA companies' business models. The market is not concerned with where the value is created, so long as value is delivered in an uninterrupted fashion.

Thus, large EDA and framework vendors should begin to accept the idea of distributing certain application software packages from third parties and phasing out internally developed products that cannot remain competitive. EDA vendors should not invest an unreasonable amount of resources in developing application software, but instead concentrate on developing only those core applications that provide them with a strategic advantage such as account control.

The EDA companies that come up with the right mixture of added value will be able to deliver the best solution, and this will provide tremendous leverage to their frameworks. This, in turn, will translate to better account control and more stable revenue streams.

In any event, the whole notion of a shortterm winner or loser in the framework wars is without basis. •

By Ron Collett

Product Analysis

Silicon Graphics

Announces Indigo, a New Low-End

Workstation for Less than $10,000

(Silicon Graphics' 33-mips Indigo workstation lowered the price tag of 3-D modeling, offering implications for CAD/CAM/CAE users on workstation platforms. In gaining immediate support from two EDA vendors, SGI's Indigo machine also staked out new claims in the 2-D graphics market for Silicon Graphics, which historically had concentrated on higher-end 3-D applications. The following article is reprinted with permission of Dataquest' s Technical Computer

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

i i i

CAD/CAM/CAE

Systems group. In addition, there is an analysis by Dataquest's Michael J. Seely on how these recent events affea the CAD/CAM/CAE industry.)

Introduction

On July 22, Silicon Graphics Inc. (SGI) announced a suite of new products. The new products include IRIS Indigo, a new low-end workstation; IRIS Explorer, a visual application environment; CASEVision, a visual development environment; and the IRISserver family of file servers. This newsletter provides a synopsis of each aimouncement and Dataquest's view of the impact of the new Indigo workstation on the industry.

Announcement

IRIS Indigo

The major produa that was announced on

July 22 was the IRIS Indigo. The system is based on a 33-MHz R3000A/R3010A, supports

8 to

9 6 M B

of memory, has up to 1.3GB of disk space, and has integrated digital audio tape

(DAT) quality audio capabilities. The system is rated at 26 SPECmarks and 4.2 mflops. Graphics performance was quoted at 250,000 2-D vectors/sec, 200,000 3-D vectors/sec, 14,000

Z-buffered flat-shaded triangles/sec, and 5,000 independent quads/sec.

A base system with 8MB of memory and a l6-inch color monitor lists for $7,995- With a

236MB formatted disk drive, the system is priced at $9,995. For the multimedia market, the typical base configuration will be the 16MB,

16-inch color monitor, and 432MB disk system, which lists for $12,500. The system will be available in volume in September.

SGI's Strategy is to expand into new application areas tvithin the technical marketplace.

An optional live video board is also available for the system. The expected availability date for the video board is the second quarter of

1992, and the board is expected to list for approximately $2,000.

IRIS Explorer

IRIS Explorer is a visual application environment that provides the end user with a means to view the numerical results of an engineering or scientific problem. The end user can simply connect various modules together in a pointand-click manner to apply various algorithms and graphics techniques to the problem without doing any programming. The package will initially include more than 100 modules. It will be available for current models and will become standard with all future systems using IRIX 4.0.

The delivery date is set for late 1991.

$26,000

Soiuce: Dataquest (August 1991)

$105,000

CASEVision

CASEVision is a program developed to address the computer-aided software engineering marketplace. CodeVision is the first product to be

Table 1 nuSserver Family Base Configurations

Data Station 2

PO\rERfile 50

35-MHz R3000

16MB memory

780MB system disk

780MB exjjansion disk

Network backup software

Ethernet

IRK 4.0/NFS

Two 33 MHz R3000

32MB memory

780MB system disk

Three 1.6GB expansion disk

Network backup software

Ethernet

IRIX 4.0/NFS

2.3GB Smm tape drive

1/4-inch tape drive

POWERflle 100

Four 33-MHz R3000

64MB memory

780MB system disk

Seven 1.6GB expansion disk

Network backup software

Dual Ethernet

IRIS 4.0/NFS

2.3GB Smm tape drive

1/4-inch tape drive

Dual VME

Quad SCSI channels

$165,000

©1991 Dataquest Incorporated / 1290 Ridder Parte Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE offered by SGI under this program. It is an integrated toolset for individual developers that includes a static analyzer, a debugger, and a performance analyzer. The CodeVision toolset will be available in 90 days, priced at $1,500.

IRISservers

SGI announced a new line of file servers: Data

Station 2, POWERfile 50, and POWERfile 100.

The systems' performance and basic characteristics are equivalent to the 4D/35, 4D/320, and

4D/340 servers but have had additional hardware, software, and networking features added to address the file server market. The base system and price for each model are shown in

Table 1. Hardware features have been added to the system, including dual POWER Channel architecture, a SCSI expansion cabinet, a 1.6GB

SCSI disk, a 1/2-inch SCSI tape drive, and a dual channel VME/SCSI controller. The new software includes network backup software, volume management software, mirroring software, network archiving software, and IRDC 4.0. All systems are currently available.

Dataquest Perspective

The most important aspect of this whole announcement is that SGI finally has a box for under $10,000. Sun Microsystems Inc. has aggressively pursued and cornered j h e low-end desktop market. Therefore, in order to compete for the desktop, a vendor has to have a box for under $10,000. Now, SGI can compete in this sector of the market. The only other leading competitors offering RISC workstations in this price range are Sun and Digital Equipment

Corporation. However, we expect new products to be announced by Hewlett-Packard Company and IBM Corporation at this price point before the end of the year. Thus, SGI has little time to capitalize on this new produa.

The second important feature is that this system is a competitive 2-D box—a first for SGI. With the majority of applications still only requiring

2-D capabilities, end users had to pay a premium for the 3-D graphics. As such, SGI was only able to penetrate accounts requiring

3-D graphics. The new system, which is competitive both in terms of performance and price with other 2-D systems available today, will broaden the total available market for SGI. And, it also puts SGI in a better position for upgrade business because it is always easier to upgrade a user to 3-D, video, or audio capabilities than to have to oust the incumbent. But more importandy, the potential for increased volumes combined with the Compaq Computer Corporation and Microsoft Corporation deals should help SGI attract new software vendors to its platform.

Additionally, SGI has an advantage over every major workstation vendor, except Digital, with its symmetric multiprocessing servers. The introduction of a server line that takes advantage of the multiprocessing as well as provides the improved functionality, such as disk backup and disk management, is a very good move on

SGI's part. According to Dataquest's Server group, file servers in 1990 represented the largest usage of servers in terms of doUars. File servers and database servers are projected to be the two largest segments by 1995.

The major drawback to this announcement was

SGI's emphasis on the audio and video capabilities of Indigo. Although these produas offer the hardware functionality, no software is currendy available that takes advantage of all these features—not even e-mail. Because of the lack of software and the faa that the multimedia market is still in its infancy, we do not expect these features to stimulate a huge increase in demand for SGI's products in the near future.

However, what audio and video capabilities will do is position the company at the forefront of multimedia technology in preparation for growth in this market.

A second area that was not addressed was the distribution channels for selling these systems.

With the lower price point, distribution through the direct channels becomes quite expensive. At this time SGI is not in a position, based on its unit volume, where it could entice other channels to offer its systems. However, it must still address the issue of distribution channels or suffer the consequences of lower profits.

In conclusion, SGI's strategy is to expand into new application areas within the technical marketplace. The price point and 2-D capabilities of Indigo will increase SGI's total available market. However, with the multimedia capabilities,

SGI may find itself actually serving more corporate commercial applications rather than technical ones because the most promising areas for multimedia are in corporate training and presentations.

From the beginning, Silicon Graphics has positioned itself as a 3-D graphics supplier, staking its claim on the high road of graphics workstations. With the introduction of Indigo, SGI has effectively brought the entry point of 2-D and

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE

3-D graphics down to a level where sophisticated 3-D modeling may begin to reach the masses.

In addition, SGI immediately garnered endorsements from the CAD/CAM/CAE market with electronic design automation (EDA) vendors

Cadence Design Systems Inc. and Valid Logic

Systems Inc. announcing p>orts to the SGI machine. With the Cadence and Valid ports, SGI can now go after opportunities in the EDA market that historically have been 2-D-only applications, such as printed circuit board (PCB) layout. Designers of electronics produas can use the Indigo and higher-end machines to operate sophisticated 2-D and 3-D electronic package design software. The ports also will eventually open up the prospect of future 3-D modeling for a variety of elearonic design problems. In the long term, even chip design may become a 3-D application with detailed design and analysis at the micron level. •

By Laura Segervall

Michael J. Seely

Highland Software

Ships Software Store

(With Highland Software's shipment of the Software Store, a collection of UNIX application software will now be available to users of

UNIX-based workstations in the CAD/CAM/CAE area. Representing a promising distribution method for CAD/CAM/CAE software, the Highland Software program has the potential to benefit both software vendors and users alike in the CAD arena. The following article, written by an analyst in Dataquest's Personal Computer

Software group, is reprinted with permission.

Additional analysis is provided by Dataquest's

Kathryn Hale on how these recent events affect the CAD/CAM/CAE industry.)

The first volume of the Software Store, a CD-

ROM containing UNIX application software packages and text and graphics exhibits from many independent software vendors (ISVs), shipped this summer from Highland Software.

Volume I of the Software Store CD contains more than 80 products from 36 companies.

Software areas covered include CAD/CAM/CAE, office automation, electronic publishing, graphics and visualization, and system utilities.

There are three versions of the Software Store— the Software Store for ULTRIX, the Software

Store for HP, and the Software Store for Sun.

Additional volumes are to be added quarterly.

Each edition of the Software Store packages 20 to 35 applications on a single CD for a single hardware platform. The editions are published quarterly.

CD-ROM technology offers CAD vendors significant opportunities to reach and influence prospective users electronically.

ISVs purchase space by the megabyte and a start-up kit that includes the following:

• Software utilities

• Documentation

• Special support for first-time exhibitors

• Beta disk for quality assurance

T h e Distribution Method

The CDs are distributed at no cost to qualified users of workstations based on the UNIX operating system, to computer company sales offices, and to computer distributors. The prospect inserts the CDs in the CD-ROM drive to examine, evaluate and, in some cases, to immediately purchase the software.

For the ISVs, the Software Store serves as a worldwide marketing, distribution, and promotion medium. It carries both demonstration and licensable, password-protected UNIX software.

The customer contacts the ISV directly to discuss purchase, removing the burden of the middleman from Highland Software. Each company informs users of the contact name for product and purchasing information.

Easy t o Use

Users insert the CDs into CD-ROM drives connected to workstations and "mount" the CD-

ROM as part of the computer's file system.

Some programs may be purchased or evaluated by contacting the ISV for a password. Installation is automatic. Each CD contains an electronic catalog of UNIX application software with password protection, demonstration versions of software, and on-line documentation. ISVs maintain control of the presentation, promotion, and distribution of the software exhibits. Highland provides a toolkit, technical support, standard software installation procedures, tools to help create exhibits, and integration of the exhibits on the CD.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

8 CAD/CAM/CAE

CD-ROMs—The Preferred Method o f Distribution?

Other companies using this method of promoting and selling their products include Digital

Equipment Corporation for its RISCAJLTRDC products, Hewlett-Packard Company for its HP

9000 family of product, and Sun Microsystems

Inc. for its SPARC computers.

The Software Store incorporates the X Window graphical user interface, allowing users to pointand-click on product demonstrations. FoUo^ving the subsequent purchase and unlocking of the software through the use of password codes provided by the software vendor, users also have the ability to personally install the software. The CD-ROM also includes text-retrieval software, which allows users to locate software by typing in key words that suggest particular applications.

An obvious qualifying tool, CD-

ROM distribution could replace both a portion of advertising in other media as well as a large number of initial sales calls.

SIOO-SISO

IL ^.t V-.

(50-5100

$1-$10

Customer Sales

Office Office

Trade Demo Software

Sfiow Tape/ Store

Manuals

Source: Highland Software

Although CD-ROM distribution unll reduce the cost of sales by automating portions of the job of selling, it also will make selling CAD software much more competitive.

Figure 1 shows the approximate costs to an ISV to promote its product through the usual methods. These costs are incurred before a prospect is qualified or any sale is made. The comparison points out very vividly the savings when using this type of method to make that first important contact.

Selling software directly is the most expensive way of getting a product to the prospect. Other methods are less costly but continue to keep the price of the software high and limit sales.

FIGURE 1

Costs per Customer Demonstration

$400-$G00

$300-$400

To exhibit on one platform on the Software

Store, charges would be $5,360 for a 3MB demonstration and $15,000 for a 25MB program, in addition to a $3,000 one-time chaiige. Shelf space on additional computer brands is available at reduced rates, and additional discounts are available no"w.

D a t a q u e s t P e r s p e c t i v e

CD-ROM technology offers CAD vendors significant opportunities to reach and influence prospective users electronically. An obvious qualifying tool, CD-ROM distribution could replace both a portion of advertising in other media as well as a large number of initial sales calls. The technology also gives the converted prospect ready access to an "internal selling" tool. We foresee an increase in political skirmishes within a prospective buyer's ranks, as internal champions of competing solutions are able to demonstrate, at will, the good and bad points of various products.

Some potential also exists to use CD-ROM delivery both to attract new dealers and to integrate resellers better with a vendor's direct sales force. With effective, interactive CD-ROM produrt demonstrations, dealers could more readily step up to front-line selling of sophisticated workstation-based software, bringing in the vendor's direct sales staff for highly qualified prosfjects. This scenario would require, of course, creativity in compensation structures.

Although CD-ROM distribution will reduce the cost of sales by automating portions of the job

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE of selling, it also will make selling CAD software much more competitive. The technology invites a "free trial," potentially exposing products to increased testing. We have long advised CAD vendors to emphasize creating products that are easy to learn, easy to use, and bug-free. We have also stressed the importance of building service, support, and training into the product itself, rather than relying on individual performers alone to supply these key services. Those vendors with products that best meet these requirements are now in a superior position to leverage CD-ROM distribution, an environment where the product that stands out is the product that performs best during a prospective buyer's uncontrolled, untutored exploration. •

By Gladys Francis

Kathryn Hale

three contingents favoring either the Apple/

Macintosh, ffiM/Intel, or SPARC/RISC/Motorola architectures for heavy-duty document imaging.

DIMS vendors, integrators, and software developers tend to support one particular ^vorkstation/server architecture for a number of diverse reasons, including performance, display capabilities, networking, multitasking, distributed computing, operating system, adherence to industry standards, price, and the preferences of their customers.

Given the fact that Sun workstations were developed with sufficient horsepower to handle scientific and technical applications, document imaging is a very low end application in the Sun universe.

Company Analysis

Workstation Vendor

Strategies in the

DIMS Market-

Sun Microsystems

We hear a lot these days about the "Worisstation Wars," technological skirmishes wherein the various workstation vendors vie for supremacy on the desktop. An ongoing battle pits Intel

PCs against SPARC/RlSC/Motorola technical worksmtions. Each camp makes claims about superior performance, compatibility, networking, and market penetration. The truth is that the distinctions between PC and technical workstation are blurring. Multitasking operating systems for the PC such as OS/2, and graphical user interfaces such as Windows 3.0 and the Presentation Manager, make PCs look and behave ever more like workstations.

But more is at stake in these wars than the desktop. Networked, distributed computing is made possible by high-performance servers that take the place of minicomputers and even mainframes. Workstation and PC vendors are now in the server business, and distributed computing has become big business.

Document image management systems (DIMS) providers cheer these battles on from the sidelines. The DIMS industry is divided roughly into

This is the first in a series of articles exploring the strategies of workstation/server vendors to position their products in the DIMS market. Our first workstation warrior is Sun Microsystems

Inc. of Mountain View, California.

Sun's P h i l o s o p h y — K e e p It

Simple, K e e p It Separate

Sun is first and foremost a manufacturer—of technical workstations, servers, and the recently introduced SPARCprinter. To assure market acceptance of these products and ensure that they operate in the most efficient manner possible, Sun's secondaty business is developing operating system software, networking software, user interfaces, and application development environments compatible across all Sun products. All Sun products are based around a single microprocessor architecture—SPARC;—thus simplifying development and support.

Sun's customers are systems integrators, large end users, and independent software vendors

(ISVs). All add value to Sun hardware by developing software to run on it, by buUding systems around it, or both. To facilitate this development, Sun provides its partners with tools including a high-level programming language, development tools, and communications software.

Sun's marketing strategy aims to position Sun products as the most high-performance, compatible, flexible, and cost-efficient available. Sun's support oi;gani2ation is chartered to provide customers with all pre- and postsale support

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-6000 / Fax (408) 437-0292

0011243

10

CAD/CAM/CAE

needed to sell and install an account successfully.

Overall, we believe that document imaging, as a horizontal application, mill be tightly integrated into desktop computing environments over the course of the 1990s.

A recent restructuring at Sun spun off two new companies chartered with developing and managing the company's investments in UNIX software development. This diversification created separate operating entities responsible for managing software (rather than hardware) development activities, as follows:

• SunSoft Inc. is chartered to manage Sun software development activities including the Sun operating system (SunOS), windowing system

(Xll/NeWS and the OpenWindows environment), and networking software supporting the Network File System (NFS) and Open

Network Computing (ONC) architecture. The group will concentrate on sales, marketing, and licensing of these technologies and work with international standards bodies to ensure that Sun software products comply with industry standards.

• SunTech Enterprises Inc. supports third-party developers with tools and products to build applications and systems around Sun hardware. Products managed by SunTech include

SPARCprinter, NeWSprint software, PC Network File System, SPARCompiler software development platform, and SunLink connectivity gateways to systems and networks.

Sun's Imaging Strategy

As the market for commercial and technical document imaging has heated up, so has Sun's interest in positioning its workstations there.

Sun has targeted five main areas of opportunity in document image management, as follows:

• Document imaging—^The scanning, storage, retrieval, and manipulation of business-size paper documents

• Engineering data management—The scanning, storage, retrieval, and manipulation of engineering drawings and related documents such as bills of material and design specifications

• Full-text search—^Locating information using software that searches for words and phrases specified by the user

• Facsimile—Sending and receiving faxes through a workstation

• Multimedia/compound documents—^Integrating a variety of media types including text, data, graphics, scanned images, video images, and voice into one on-line "document"

Sun's Imaging Product Line

Sun's main product line consists of a family of workstations and servers built around the

SPARC microprocessor architecture, running under the UNIX operating system, and adhering to the client/server computing model.

Sun appears to be in the right place at the right time with a ivellengineered solution for bringing general-purpose document imaging to the UNIX desktop.

Sun's original products were high-end technical w^orkstations designed for use in scientific and engineering applications. To take advantage of emerging applications for workstations in the commercial market, over the past 18 months

Sun has introduced a series of lower-cost, desktop workstations aimed at business customers.

Base prices for Sun's high-end workstations, typically used for technical applications, range between $29,900 and $69,900. By contrast, lower-end workstations for the commercial market cost much less. The diskless SPARCstation

SLC sells for a base price of $4,995, and the

SPARCstation IPC sells for $7,995. Base prices for the four servers in the SPARCserver family range from $24,590 for the SPARC:server 2 to

$99,900 for the SPARCserver 490.

In September 1990, Sun introduced the SPARCprinter, a 12-pages-per-minute (ppm) desktop laser printer powered by Sun NeWSprint software. Newsprint builds on Sun's Network

Extensible Window System and offers PostScript compatibility. Printing has always been something of a bottleneck on Sun's otherwise highperformance systems. By solving this through a software solution, Sun users can take advantage of the liigh throughput potential of the SPARC architecture. The printer and software are bundled for a list price of $2,695-

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE 11

The company believes that its SPARCstations and servers are ideal platforms for handling document imaging applications, citing the following reasons:

• Performance—^Technical workstations generally offer high-performance capabilities required to run image-processing appUcations, such as: fast processing speeds; extended memory; multitasking; ample disk storage; highresolution, large-screen displays; and built-in networking capabilities.

• Networking—Server-based document imaging networks can be set up to grow as the user's needs grow, to support other applications in addition to document imaging, and to tie into other computer systems and databases within the corporation. In addition,

Sun's ONC architeaure ties computers from other vendors into the Sun environment.

• Distributed computing—The client/server model provides an efficient style of computing in DIMS environments—^if one server goes down, other functions can proceed uninterrupted.

• Database management—^High-performance servers are well suited to handle image database management funaions in installations comprising large numbers of image-intensive workstations.

• Multitasking—Sun workstations combined with the UMDC operating system enable users to perform several tasks concurrently on a workstation—^for example, to simultaneously view an image, pull in c^ta from a mainframe database, and integrate this information into a word processing document.

• Mukimedia—Sun workstations arc designed to display and integrate mixed data types including text, data, graphics, scanned images, video images, and voice.

• Price—Costs are declining to the point where

Sun workstations are competitive with Intelbased PCs often used in image-processing environments (see Table 1).

Sun's Imaging Partners

Sun relies on systems integrators and software developer partners to integrate its workstations and servers into document imaging installations.

The integrators usually work as prime contractors, pulling Sun into the bid if the customer's installation warrants a Sun-type solution. In the case of software developers that port software to Sun platforms, sales may be driven by the software partner, an integrator using the software solution, or Sun's sales force.

Sun works with leading integrators to install commercial and technical DIMS. Sun's commercial DIMS integrators include Andersen Ckansulting, Electronic Data Systems (EDS), FileNet,

Table 1

SPARCstations versus PCs

Configuration

Standard

PC

200 MHz,

386/2MB,

RAM/60MB disk

Imaging PC

SPARCstation 1 Production

SLC

Imaging PC

25 MHz, 386/4MB,

RAM/60MB disk

20 MHz,

SPARC/8MB,

RAM/diskless

SPARCstation IPC

33 MHz, 386/8MB,

RAM/IOOMB disk

25 MHz, SPARC/

8MB, RAM/

207MB disk

Lisl/Street Price ($)

Mips

Monitor

Price

Monitor Controller/Driver Price

5,600/3,500

4.5

6,800/3,500

6.2

6,500

8

14" VGA, 57 dpi

14" Super VGA, 70 dpi

17" 1152 X 900,

100 dpi

19" high resolution, 115 dpi

Included

Included

$450

Included

4,995

12.5

Included

Included

$2,050

$1,900

7,995

15.8

OS/Windows Price

Cable Price

Image Adapter Board Price

Network I/F Price

Mouse Price

Total

Included

Included

NA

$450

$100

$4,100

$270

$155

$950

$450

$100

$5,875

Included

Included

NA

Included

Included

$4,995

$270

$155

$1,500

$450

$100

$12,925 $10,695

NA = Not available

Note: Street prices are generally unavailable on production imaging PCs because of tbe need to work with an integrator.

Source: Sun Microsystems Inc., Dataquest (August 1991)

19" high resolution, 115 dpi

$2,700

Included

Included

Included

NA

Included

Included

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

12 CAD/CAM/CAE

Genesis Imaging Technologies, Grumman

InfoConversion, NYNEX, Perot Systems, Philips

Information Systems, Price Waterhouse, TRW

Financial Systems, a n d Xerox. Sun's technical

DIMS integrators include Cimage, FORMTEK,

GTX, Litton/Integrated Automation, Optigraphics, and Planning Research Corporation (PRC). Table 2 lists leading software partners that have ported d o c u m e n t imaging solutions to Sun platforms.

(Sun's partners place an order for a certain n u m b e r of workstations and servers, but d o not necessarily tell Sun w h e r e they are going), Sun recently listed 77 d o c u m e n t imaging installations in place w o r l d ^ d e using Sun workstations or server/workstation combinations. Approximately

62 of these systems are installed in the United

States, 13 in Europe, and 2 in the Far East.

Imaging Installations

T h o u g h the exact n u m b e r of imaging installations running Sun hardware is difficult to track

Table 2

Sun's I m a g i n g Software Partners

Partner

Calera Recognition Systems

JTS Computer Systems

NYNEX Image Recognition Systems

OCR Systems

Plexus

Resumix

Workgroup Technologies

Xerox Imaging Systems

Soiiice: Sun Microsystems Inc.

As s h o w n in Figure 1, t h e majority of Sun's reported document imaging sites are in the manufacturing sector. Users in this industry include Boeing, Ford, GE Aircraft Engines, Mack

Product Ported to Sun Environment

OCR software and developier's kit

OPENIMAGE document management solution for papier-intensive environments

Intelligent character recognition software that reads handprinted characters

ReadRight OCR software

Plexus XDP image-processing software on Sun platforms

Resumix resume-processing and piersonnel management software

Technical data and document management software

ScanWorX text recognition and image capture system

Flgiure 1

Sun's D o c u m e n t I m a g i n g Installed Base b y Industry

Telecommunications

Utilities 10%

Pharmaceutical/

IHealth Care 8%

Legal 6%

Financial Services 4%

I

i

Source: Sun Microsystems Inc.

©1991 Dataquest Incorporated / 1290 Ridder Parit Drive, San Jose, CA 95131-2398 / (408) 437-SOOO / Fax (408) 437-0292

0011243

i

CAD/CAM/CAE

13

Truck, Martin Marietta, Pratt & Whitney, Rockwell Space Division, and Rolls Royce. Government sites comprise 18 percent of installations, with major U.S. federal customers including the

Department of Defense, the Environmental Protection Agency, the Library of Congress, the

U.S. Navy, and the U.S. Secret Service; and dty government customers including the City of

Chicago's Bureau of Parking and the New York

City Board of Elections. Insurance and banking account for 12 and 11 percent of Sun installations, respectively. Major bank customers include Bank of America and Citicorp Mortgage.

Insurance providers include Blue Cross/Blue

Shield, Great Western Life, and New York Life.

Major customers from other industries include:

Merrill Lynch and Fidelity Investments (financial services); the University of California, Berkeley

(education); J.C. Penney, Northwest Airlines, and

Walt Disney Pictures (general business); Bristol-

Meyers Squibb and Merck, Sharp and Dohme

(pharmaceuticals); and AT&T, British Telecom, and Pacific Bell (telecommunication).

The Bottom Line

Document imaging is a steadily growing business at Sun. In Sun's 1990 fiscal year, sales of workstations and servers into document imaging installations accounted for $100 million of Sun's total revenue. In fiscal 1991, this number had grown to $140 million, an increase of 40 percent. This represented 5 percent of Sun's total revenue for fiscal 1991 (roughly 3 percent was commercial DIMS and 2 percent was technical

DIMS).

Dataquest Perspective

Given the faa that Sun workstations were developed with sufficient horsepower to handle scientific and technical applications, document imaging is a very low end application in the

Sun universe. If anything, Sun workstations are probably "overqual^ed for the job." But their fast [performance, multitasking capabilities, and integrated networking, combined with increasingly attractive list prices, make them evermore popular choices for managing DIMS databases.

Judging from the distinguished list of partners that work with Sun to provide document imaging solutions, the Sun workstation/server model is becoming an industry standard among

UNIX-based DIMS providers.

We believe that Sun will successfully further penetrate this market for the following reasons:

• Consistent quality of product

• Ever-increasing price/performance ratios

• Dependable OS/networking software services to support the Sun client/server model

• Ability of Sun systems to reside peacefully and share information in multivendor environments

• Sun companies chartered to develop and support third parties and ISVs

• Support organizations within Sun focused on customer service

• Consistent focus from the corporate level down toward development of a single, integrated product line

Overall, we believe that document imaging, as a horizontal application, will be tightly integrated into desktop computing environments over the course of the 1990s. Systems offering superior networking and the ability to handle a variety of compound data types will lead this movement. Sun may well be at the vanguard of this trend. Its competitors face unresolved problems: The Intel PC crowd dithers over the acceptance of OS/2; proprietary systems vendors struggle to open up closed architectures, support multiple platforms, and move to UNIX at the same time; the Macintosh provides applications consistency and gorgeous graphics capabilities, but so far is still a relatively proprietary solution and has achieved low penetration of the general business and technical markets. Sun appears to be in the right place at the right time with a well-engineered solution for bringing general-purpose document imaging to the

UNIX desktop.

One note of caution: The recent restructure that spun off the SunSoft and SunTech companies has left gaps to be filled. Many employees find themselves not only working in new jobs, but at new companies as well. In some instances, replacements have not been identified or may not be hired, leaving gaps in the previous infrastructure. This restruaure affeaed the marketing team for Sun's commercial document image systems development, for example, but left the TDIMS team relatively intact. Although the formation of SunSoft and SunTech will promote better long-term support for Sun customers, we expect it to take some months before business returns to normal. •

By Pamela Stone Bliss

(This article is reprinted with the permission of

Dataquest's Document Imaging service.)

©1991 Dataquest Incorporated / 1290 Rldder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

14

CAD/CAM/CAE

News and Views

Cadence and Siemens

Nixdorf Enter Joint

EDA Venture

Cadence Design Systems Inc., of San Jose,

California, and Siemens Nixdorf Informationsysteme AG (SNI), of Paderborn, Germany, last month armounced their intention to combine operations in central Europe to jointly develop, market, and support EDA solutions for the

European market. The joint venture's headquarters will be in Munich, Germany, and will include the combined sales, marketing, and service oiganizations of the two companies.

The new company, called Cadence Europe

GmbH, will be jointly owned by Cadence and

Siemens Nixdorf, with Cadence holding a oneshare majority stake. The combined organization aims to leverage the technology and expertise of the companies in order to provide a set of products based on a unified framework,

Cadence's Design Framework 11.

The Cadence Perspective

Dataquest believes that Cadence will accrue the following benefits from this joint venture:

• Through this deal. Cadence potentially gains greater access to Europe's biggest EDA market: Germany, which accounted for 33 percent of all EDA sales in Europe in 1990. In the longer term, entry into Eastern Europe will be eased due to Germany's (and Siemen's) special links with the East.

• Cadence positions itself as the champion of open framework standards. Its association with Siemens, a leader in the Jessi European framework standards effort, will enhance its image as a key player in the framework scene.

• Cadence has moved from a niche IC layout company to embrace CAE and printed circuit board (PCB) layout. SNI owns key technology—^particularly in the systems, CAM, testability, data management, and EMC areas.

If properly managed, this technology could enhance Cadence's product offerings considerably, moving the company further into the broad electronic product development field.

The Siemens Nixdorf Perspective

By joining forces with Cadence, SNI gains the following:

• Assistance in the marketing and support of products within the German region

• Accelerated returns on investment in EDA tools (The rate of returns following the Calay

Systems Inc. acquisition and substantial internal development was proving slow for SNI.)

• A ready-made, credible product line for sales into the Siemens organization (Siemens' increasing need for IC development tools is likely to have been a major factor in this respect.)

Dataquest Perspective

This deal follows Cadence's joint venture with

European Silicon Structures last year, when a

European EDA development and marketing center was established in the United Kingdom.

Cadence is clearly strengthening its underdeveloped distribution channels in Europe (in comparison with other broad-line EDA vendors) through a strategy of joint venture and direa channels to suit local market conditions.

The deal gives Cadence greater access to the

SNI organization, which is an appreciable slice of the German market. However, SNI does not operate a 'single-vendor' EDA policy, and

Cadence will have to fight alongside other suppliers—^notably Racal-Redac Inc., Dazix Intergraph, Mentor Graphics Corporation, and Valid

Logic Systems Inc., all of which have sizable user bases within SNI. The deal is also a valuable springboard into the wider German EDA market, which was valued at $318 million in

1990, using Cadence's strengths in other geographical regions as an additional credibility factor.

From SNI's viewpoint, the following possibilities exist:

• SNI realized it could not survive in the EDA business in view of its late entry and, therefore, has effectively sold out to Cadence. We do not believe this to be true. More likely,

SNI is serious in its intentions to become a world EDA player. SNI usually takes a longterm strategic view, and the company would not have acquired the assets of Calay Systems last year if it was not serious in its intentions (although the future of Calay itself is now open to speculation).

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

CAD/CAM/CAE 15

• SNI found itself in a vulnerable position because many of its products are sourced from competitors. The company did not have the funds to develop or acquire replacement products, and, therefore, teamed up with a company that had a strong and urgent need to increase its share of the lucrative German market. This latter point places SNI into a strong bargaining position.

• SNI needed to extend its operations on a wide geographical base. The company decided to achieve this by tapping into the technology and culture of Cadence, one of the rising stars of the EDA world. In this scenario, one thing does not add up: If the venture covered a different geographical region (for example. North America or Asia),

SNI would be gaining access to markets that are essential if it wishes to become a world player in EDA. However, the agreement is restricted to the geographical region around

Germany—SNI's home territory. We believe that there must be more in this for Siemens

(such as access to Cadence's wider distribution channels) and that further announcements will follow.

In the short term, this agreement will have little effect on the European EDA market. In the longer term, we believe that Cadence's position in Europe will be enhanced by it. One of the most interesting questions relates to the culture of the two companies: Will the fast-moving, dynamic Cadence and the large, slower-moving

SNI result in a happy marriage, or, instead, will there be a serious personality dash? •

By Jim Tully

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-«000 / Fax (408) 437-0292

0011243

16

In Future issues

Topics of interest to the CAD/CAM/CAE industry will appear in upcoming issues of Dataquest

Perspective, including the following:

• A report on the mechanical applications market

• An overview of the European electronic design automation (EDA) market

More "News and Views" on the CAD/CAM/

CAE industry

CAD/CAM/CAE

i i

For More Information » . .

Managing Editor Patrick Waurzyniak (408) 437-8132

On the topics in this issue CAD/CAM/CAE (408) 437-8132

On related subjects Client Inquiry Center (408) 437-8671

About other Dataquest publications Sales (408) 437-8000

About upcoming Dataquest conferences Conferences (408) 437-8245

About your subscription Customer Service (408) 437-8402

Via fax request Fax (408) 437-0292

The content of this report represents our interpretation and analysis of information generally available to the public or released by responsible individuals In the subject cofi^>anic9, but is not guaranteed as to accuracy or comF^eteness. It does not contain material provided to us in confidence by our clients. Individual companies reported on and analyzed by

Dataquest may be clients of this and/or other Dataquest services. This informadon is not furnished in connection with a sale or offer to sell aecunties or in connecuon with the solicitation of an offer to buy securities. This firm and Its parent and/or their officers, stockholders, or members of their fomilles nuy, from time to time, have a long or short position in the securities mentioned and may sell or buy such securities.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0011243

i

Datac^est

n n a company of

MMMB TheDun&BradsticctcorpwaDon

Dataquest

Perspective

CAD/CAM/CAE

Vol. 1, No. 2 July 8, 1991

Market Analysis

A View from 15,000 Feet: Looking M the Landscape of the Entire CAD/CAM/CAE Industry

Following up on our recent overview of the CAD/CIAM/CAE world versus the entire worldwide computing base, Dataquest presents its final 1990 market share data for the CAD/

CAM/CAE industry and takes a closer look at a CAD industry that grew nearly 15 percent to reach the $14.4 billion level in overall revenue during the 1989 to 1990 time frame.

By CAD/CAM/CAE industry service staff

Page 2

Product Analysis

ComptOervision's CADDS 5: More them Just a Face-Lift

Computervision's first major update of its software in six years takes its new computeraided design and manufacturing application well beyond the company's previous offering,

CADDS 4. Dataquest takes a hands-on look under tiie hood of CADDS 5, which includes a completely reengineered user interface as well as several enhancements of the software's design and modeling capabilities.

By Michael J. Seely Page 4

Technology Analysis

What Windows Means in the Long Term

Microsoft Corporation's popular Windows graphical user interface has generated some heated discussions, but these arguments usually center more on short-term issues such as marketing agreements and not on the product's lasting impact on the personal computing world. Dataquest examines Microsoft's long-term Windows strategy and its probable effects.

By Marshall L. Moseley

Page 7

News and Views

An analysis of recent news events affecting the CAD/CAM/CAE industry

Hewlett-Packard to Port Mechanical-Design Software to Sun SPARCstation

ESRI Goes GUI

Intergraph's Dazix Subsidiary Signs Technology Pact with AT&T

Page 10

Page 10

Page 11

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

CAD/CAM/CAE

Market Analysis

A View from 15,000

Feet: Looking at the

Landscape of the

Entire CAD/CAM/CAE

Industry

Analysis of Dataquest's final 1990 market share update shows that the entire worldwide CAD/

CAM/CAE market grew at a 14.6 percent rate from 1989 to 1990, a period in which the industry rose in value from $12.5 billion to

$14.4 billion (see Figure 1 and Table 1).

The largest application area among individual

CAD/CAM/CAE market segments shown by market size and market growth rate is mechanical applications, valued at almost $7.5 billion,

JF^Iure 1

Worldwide CAD/CAM/CAE Industry Maricet Portfolio

1989 to 1990 Market Growth

(Percent)

35

which lies almost directly on the industryaverage growth line at about 15 percent. The

PCB/Hybrid/MCM and Electronic CAE applications areas, each of which is a much smaller market segment overall than is the mechanical market, also grew at a slower-than-average rate.

In contrast, the GIS/Mapping market continues to be the CAD/CAM/CAE industry's fastestgrowing application area with a growth rate of nearly 30 percent, a rate significandy higher— almost double—than the industry average.

Technical workstations, by far the largest platform based on CAD/CAM/CAE revenue, grew at an above-average rate in 1990, albeit at a somewhat slower rate than it had in 1989. The revenue spent on technical workstation-based systems now totals more than that of personal computers and host-based systems combined.

Although technical workstations dominate the current sales picture, personal computers hold the largest share of installed base, with just over 1 million shipped by the end of 1990.

This compares widi about 401,000 technical workstations and 300,000 host-based terminals shipped.

30

25

GIS

Total 1990 CAD/CAM/CAE Market = $14.4 Billion

20

15

AEC

Industry

10

5

,PCB) V r V 14.6%

"I Host

Dependent

-r

3

6

Market Size (Billions of Dollars)

Note: The size of each bubble reflects the size of the market. The bubbles are centered over both their market size and their market growth rate last year.

The sum of any one shading style equals the total $14.4 billion CAD market.

Source: Dataquest (July 1991)

8

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-6000 / Fax (408) 437-0292

0010869

1

9

CAD/CAM/CAE

3

Table 1

W o r l d w i d e CAD/CAM/CAE Maxket

IBM

Intergraph

Computervision

Digital

Hewlett-Packard

Sun

Mentor Graphics

Compaq

Fujitsu

McDonnell Douglas

NEC

Siemens

Autodesk

Apple Computer

Cadence

Nihon Unisys

Hitachi

Control Data

Valid

Silicon Graphics

Others

All Companies

All U.S.-Based Companies

All Asia-Based Companies

All Europe-Based Companies

All Hardware Ck>mpanies

All Turnkey and Software Companies

Total

Revenue

($M)

1,777.3

1,013.3

954.1

904.6

812.4

210.9

208.3

178.2

158.5

151.3

4,549.0

14,380.4

10,887.3

2,047.3

1,445.6

3,949.6

10,430.9

Note: Columns may not add to totals shown because of rounding.

Sonice: Dataquest (July 1991)

589.3

425.2

420.5

395.4

352.7

342.4

269.7

237.8

218.4

211.2

This snapshot of the CAD world, provided in order to give a capsulized view of the 1990

CAD/CAM/CAE market, is available in presentation form upon request.

Market Database Methodology

Dataquest's CAD/CAM/CAE industry service is committed to consistendy delivering to clients the most accurate data available, both in terms of tnarket history and forecast. Our annual delivery schedule is designed to meet that commitment, and is as follows:

• First quarter January 31): Complete preliminary market share tables for the year just completed. The historical database is opened for changes for a six-month period.

Software

Revenue

($M)

343.6

201.2

265.4

173.1

27.4

87.5

24.5

90.6

0

1,969.2

4,080.1

2,745.7

604.2

730.2

0

4,080.1

3.9

75.9

0

178.6

0

107.4

122.0

77.0

95.2

237.8

0

Second quarter (March 31): Complete preliminary forecast tables, including a new fiveyear forecast period.

Third quarter (July 1): Complete final market share tables for the previous year, based on additional data collection and analysis of the previous six months. At this point, the CAD database is frozen and will not be changed until the end of the year. During this time, supplementary market data will be based on this data set.

Fourth quarter (Oaober 1): Complete final forecast tables, taking into consideration changes in the market during the previous six months.

Software

Revenue

Share

(%)

8.4

4.9

6.5

0.1

0.6

2.2

0

48.2

100.0

1.9

2.3

5.8

0

4.2

0.7

2.1

1.9

0

4.4

0

2.6

3.0

67.3

14.8

17.9

0

100.0

Total

Revenue

Share

(%)

12.4

7.1

6.6

6.3

5.7

4.1

3.0

2.9

2.8

2.5

2.4

1.9

1.7

1.5

1.5

1.5

1.4

1.2

1.1

1.1

31.6

100.0

75.7

14.2

10.1

27.5

72.5

©1991 Dataquest Incorporated / 1290 Fidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

CAD/CAM/CAE

Since January, when the preliminary market share numbers were published, more than 35 companies have been added to the market share database, bringing the total number of companies recognized in the CAD/CAM/CAE market to 430. The market share numbers also have been rechecked and verified for all vendors worldwide. Year-end 1990 financial results released this past spring have also been checked and incorporated into the final numbers.

A number of companies significantly improved or made changes to their reporting methods during this past update cycle. For example,

Intergraph Corporation instituted new internal procedures that allow the company to more accurately track its hardware revenue by application. liiis change has resulted in a restating of Intergraph's application market share numbers for several years.

In addition, Computervision, a Prime Computer

Inc. subsidiary, also improved its internal accounting procedures to more accurately track the company's CAD/CAM/CAE revenue since the merger with Prime in 1988. As a result, Computervision's market share numbers since 1988 have been restated. Another significant change was the addition of worldwide revenue from

Europe-based ESRI Inc. offices, which are

49 percent owned by the Dangermond family.

Dataquest's policy is to continually update its market information, for current and past years, with any new data received in order to arrive at the most accurate market representation possible.

Final market share tables are currently being published in the new Source: Dataquest format.

Dataquest Perspective

The market dynamics in the total CAD/CAM/

CAE market are comprised of factors based on application-specific market needs, diverse geographical issues, and the ama2±ng growth of computing and graphics resources. A more detailed analysis of each application area must be made to thoroughly understand the forces driving the entire market. Dataquest will discuss these forces in future issues of Dataquest Per-

spective and will provide an in-depth analysis of

CAD/CAM/CAE regional issues, technology trends, and company analyses.

Dauquest's analysis of the total CAD/CAM/CAE market by platform is as follows:

• Technical workstations will continue high revenue growth with unit shipments matching personal computer unit shipments by 1995-

• Personal computers will retain a 70 to

75 percent of installed-base seats in the next five years.

• Host-based systems will remain an important faaor in the CAD/CAM/CAE business, although revenue and terminal shipments are expected to decline from the peak in

1990. •

By CAD/CAM/CAE industry service staff

Product Analysis

Computervision's

CADDS 5: Mcn-e than

Just a Face-Lift

Computervision, a Prime Computer Inc. subsidiary based in Bedford, Massachusetts, recently introduced a new computer-aided design and manufacturing (CAD/CAM) architecture that significantly enhances users' productivity and speeds the product design process. The new product, CADDS 5, incorporates advanced technology—^parametric modeling, variational geometry, a sketcher, an intelligent user interfece, a constraint modeler, feature-based modeling, and X Window access—an open database architecture, and a comprehensive suite of integrated applications.

Pticing/Packaging

CADDS 5 has been packaged with a simplified sales strategy for the benefit of both salespeople and purchasers. The basic upgrade price from CADDS 4X on a Sun SPARCstation is zero dollars through the end of the year. Those upgrading from CDS and Motorola-based hardware to SPARC-based workstations can also get the zero-dollar upgrade deal. Pricing for the whole suite of modules has been redone. The

"core" or Premium Engineering Package is priced at $24,500, with additional seats priced the same with corporate volume discounting.

Figure 1 shows Computervision's pricing schedules for the core and new modules offered in

CADDS 5.

The five standalone, task-oriented modules are interoperable and are designed to fit the needs of a certain user. All are shipped on CD-ROM with a network license manager to keep track of the number of simultaneous users. Design-

View is a variational sketcher separate from the

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

CAD/CAM/CAE

"sketcher" in CADDS 5. This module is an extension of the product developed by Premise that was recently acquired by Computervision.

DesignView is aimed at the design engineer for conceptual design in a 2-D mode with a powerful equation solver, links to spreadsheets, and a binary database compatibility with

CADDS 5 (in future releases).

View & Markup is targeted at detail checkers, engineering management, and purchasing agents. This module gives direct database access for viewing, measuring, annotation, and markup.

Design & Drafting is a standalone 3-D wireframe design tool with MELSPEC dimensioning and hidden line removal. It can function as an integrated, downstream detailiag tool for surface and solid models. Solid Modeling is a precise NUKBS-based boundary representation solid modeler. For $6,000, it offers an integrated, full function, design modeling tool.

Parametric Design drives all geometric entity types in a multiple component design environment. This module leverages the value of the new as well as old CADDS design files. Users can overlay new variational parameters on old geometry. Premium Engineering Package is targeted at the power user and multiapplication

Figure 1

CADDS 5 Priciag and Packi^ing

designers and engineers. It offers a complete set of parametric wireframe, surface, and solid modeling construction tools. It also includes a variational geometry/equation solver, featurebased modeling, and a sketcher. It will also operate with multiple-component designs and offer transparent access to any licensed application option.

The Demo

Systems engineers should receive gold medals and hazardous-duty pay for demonstrating beta software to industry analysts and current users.

The demo always dies somewhere along the way, usually just after a silent prayer has gone unanswered. A recent CADDS 5 demonstration was a better experience than most, but a few rough edges still were visible.

Among the common statements uttered during our CADDS 5 demo were "no surprises" and "a lot of functionality." The "no surprises" referred to the new user interface, which was complete, intuitive, easy-to-maneuver, and predictable. The structural aspects of the user interface seemed well conceived. The issue of functionality became more evident along the way. Each major function and subfunction would cascade

CADDS 4X Upgrade $9,800.

Waived for SPARC/CADDS through Dec. 3 1 , 1991.

Premium

Engineering Package

Any/all

MCAE/CAD/CAIVI applications

-"Core

Interoperable application extensibility-conventional licensing

DesignView

$3,000

V i e w &

Markup

$3,000

Interoperable (lOP) Series

Standalone, task-oriented modules

Interoperable-no monthly license fees

Network license management

Any module available from any CADDS seat, priced per simultaneous user

Source: Computervision

©1991 Dataquest Incorporated / 1290 Eidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

CAD/CAM/CAE

into a long, well-oiganized series of menu options with a slight movement of the mouse.

A built-in customizing option will keep customer system engineers happy by allowing full access to the look and feel of the interface.

The aptly named Sketcher anticipates an easyto-use, free-flowing graphics tool. Tlie inferencing features incorporated into the product go way beyond the expectations of a simple sketching operation. Every operation has a liveaction interpretation of every possible geometric constraint. Lines, arcs, and curves snap around the screen with dynamic text labels calling out the status of the proposed construction. Tangent, midpoint, endpoint, and perpendicular locations are all flashed in real time. The user just has to point and dick at the right moment in order to capture the relationship desired.

Later scaling or editing is accomplished, as advertised, with "no surprises," as the user has total freedom to add or remove constraints in any order. The resulting sketch is saved as an entity that can be immediately used with further geometric operations for surfacing or solidmodeling tasks.

Feature-based modeling comes with the required array of counter-bored and threaded holes. The variety of features can be placed with the wide array of geometric construction position tools incorporated in the system. A circular bolt hole pattern, ^^rhich can be generated easily with many options for starting point and angle, is a good example. A small option for through holes—with automatic update if the part thickness is later changed—exemplifies Computervision's claims of "a lot of functionality" for

CADDS 5. This seemingly small feature can save many headaches for a designer working under a time crunch.

The detail drafting menu was not available in this beta version. However, if the same attention is given to development in this important area, one would expect the delivery of a complete and flexible drafting operation. The level of automatic update from the 3-D design model remains to be seen.

Although it was difficult to find something to complain about during the demo of CADDS 5, there were a few glitches. For instance, the system once bounced back to the operating system without warning. Keeping in mind that this is beta software, one must remember that the legions of C and FORTRAN code are working hard behind the scenes to make it all happen.

The good news is that millions of lines of code written and debugged over years of development are at the user's fingertips. The bad news is the performance. A simple, trim operation of passing a bent wire through a solid of revolution took 30 seconds. Granted, this beta version of code has not been optimized for speed, and users can also expect

SPARCstation performance to improve dramatically in the future. Another caveat is the realization that this operation would have taken a talented systems jockey an hour or two with the use of 1989-vintage geometry construction tools.

Overall, CADDS 5 is a major enhancement sure to be greeted with enthusiasm from new and seasoned system users. The fact remains, however, that ftill-function CAD/CAM/CAE systems with state-of-the-art user interfaces need h^h-

end computing and graphics resources to be effective. This is true for Computervision and every other major systems integration vendor.

The reality of the situation is that if the hardware doesn't make the application sing with

"interactive" performance, it is perceived as slow, boring, and/or nonproductive. A Formula

1 race car with a two-cylinder engine gives an unfair advantage to the competition.

Dataquest Perspective

CADDS 5, ^vhich Computervision announced on

June 24, was the company's first major product release in more than five years. It incorporates a complete reengineering of the user interface, along with significant enhancements such as feature-based modeling, an inference-based sketcher, and constraint modeling.

In order to complete CADDS 5, which includes a rewrite of an estimated one-third of the code in its core software, Computervision utilized the efforts of more than 50 software programmers over a period of more tfian two years. About

45 percent of this code created the new user interface.

First customer shipment of CADDS 5 is due in

Oaober, or about 90 days after the product's introduction. Computervision is working diligently on quality, because the company wants

CADDS 5 to be a home run in the eyes of the worldwide installed base. An indication of Computervision's efforts is the significant number of companies participating in the beta test program. More than 75 customers are actively reporting bugs and suggested enhancements.

Computervision expects to leverage momentum from a rapid success here and seek other Sunbased mechanical engineers and designers. •

By Michael J. Seely

©1991 Dataquest Incorporated / 1290 Kidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

CAD/CAM/CAE

Technology Analysi:^

What Wimknvs Means in the Long Term

Reading b e t w e e n t h e Lines

(Note: Over the past year, Microsoft Corporation's Windows 3-0 graphical tiser interface has taken the personal computing world by storm, rapidly emerging as a de facto standard for graphical operating environments for the PC world as Windows upstaged the OS/2 operating system. Over the last few months, more and more CAD vendors have introduced Windowsbased CAD packages or announced plans to do so within the next year. The following article, written by an analyst in Dataquest's Personal

Computer Software service, examines Microsoft's long-term strategy for Windows and forecasts

Windows' future impact on OS/2 and Presentation Manager.)

Much of the furor over Microsoft (Corporation's wildly successful Windows graphical user interface (GUI) has concerned short-term issues, such as marketing agreements, and not its lasting effect on personal computing. Lost amidst all the recent Windows hype is a dear understanding of Microsoft's long-term strategy for

Windows. That strategy is one in which Windows dominates not only DOS platforms, but every new platform that comes into existence.

An Unexpected Success

The original version of Windows, first released in 1985, was not a popular product. Microsoft upgraded Windows in 1987, after which the product was moderately successful. Ctertainly no one expected Windows to do better than it was doing in late 1988. Most research organizations

(including Dataquest) and trade press predicted that sales of Windows would fall off once

OS/2, the next-generation operating system from

Microsoft and IBM, became available. But when

OS/2 was released, it failed to succeed as its creators hoped.

Then, in May 1990, Windows 3.0 came to market, and its sales shot through the roof. Windows 3 0 sold over 2 million copies in 1990.

Dataquest estimates that 5 million units of windows will be sold by this year's end.

No one was prepared for this degree of success, especially Microsoft. In a period of less than a year, Microsoft had to completely redesign its operating system and GUI strategy to take advantage of tiie Windows phenomenon.

That was a problem because Windows 3.0 was not designed to be the foundation of a longterm strategy. OS/2's GUI, the Presentation Manager (PM), was supposed to be the interface of the future. Nevertheless, the market was saying dearly that it wanted Windows and DOS, not

OS/2. So Microsoft went back to the drawing board and came u p with its New Technology

(NT) strategy.

New Technology Excludes

Presentation Manager

New Technology is a repositioning scheme that has diaracter-based DOS as the extreme lowend, entry-level system. It places Windows as the new midrange system and OS/2 3.0 as the high-end system. Character-based DOS, as explained by Microsoft, will continue to be supported and enhanced, but the strategic longterm goals of the company are oriented around

Windows and OS/2.

Microsoft now describes Windows as a midrange product that will exist in two incarnations: Win-l6 and Win-32. Win-l6 is the current

Windows 3.0. It is written to execute on 8- and l6-bit microprocessors. (The current Windows does make use of some 32-bit 80386 operating modes, but only to effectively multitask l6-bit

DOS-based applications.) Over time, the Windows API will be enhanced with 32-bit functionality, until eventually it will be a full 32-bit environment that can execute under DOS and

OS/2.

Another way to say this is that Windows will be available as an alternate interface to OS/2, existing with or replacing the Presentation

Manager. With this announcement, Microsoft departed completely from previously stated plans, sfjedfically the strategy outlined at the

Microsoft System Seminar in February 1989. At that seminar, Microsoft stated that OS/2 and the

Presentation Manager (PM) were its only strategic long-term operating system and GUI products. With NT, Microsoft not only brought

Windows into the equation, it positioned

Windows squarely in the center of its longterm plans and gave it equal status to the

Presentation Manager.

Reworking its operating system strategy to encompass Windows has strained one of

Microsoft's most valuable assets: its relationship with IBM.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

8

CAD/CAM/CAE

The IBM Relationship

In the mid-1980s Microsoft and IBM thought that neither could live without the other.

Together, and almost by accident, they created the IBM PC computing standard, which gave rise to a multibillion-dollar market. IBM's hardware and Microsoft's software were two equal, essential parts of that standard. When they began the OS/2 development effort in 1986, both thought that as a team they would use the new system to usher the PC user community into the 1990s. quite the same as in the past. From Microsoft's perspective, IBM refused to face reality. From

IBM's perspective, Microsoft abandoned its commitment when the going got tough. Each company is now treating the other with caution and is formulating its product strategies with less regard for its partner.

Whither Windows?

The outlook for Windows running on a DOS system is a rosy one. Dataquest estimates that sales of DOS systems running Windows will outnumber sales of non-Windows DOS systems by 1993 (see Figure 1).

Microsoft has grotun to the point that it thinks it can do some things that directly conflict unth IBM's wishes.

The long-term implication of the portable Windows strategy is that, as time goes by, operating systems will become less of an issue.

Things didn't work out that way. In its first two years on the market OS/2 failed to meet sales expectations, despite aggressive promotion from

Microsoft and IBM. This shortfall occurred primarily because OS/2's hardware requirements were prohibitive, powerful applications did not arrive on the market when scheduled, and because IBM and Microsoft failed to deliver key components of the system, such as printer drivers, in a timely manner.

When OS/2 failed to sell on desktop PC:s, IBM and Microsoft came to a parting of the ways.

As OS/2's poor sales became evident, Microsoft began to position it as more of a server-based platform than a desktop system. IBM meanwhile continued to promote OS/2 as a desktop system. The companies attempted a recondliation at Fall COMDEX, wherein they specified what types of systems should run Windows and

OS/2. Then Windows 3-0 debuted and started selling well, and Microsoft announced the NT strategy.

The NT strategy alienated IBM even more. As a result of NT, IBM and Microsoft have desktop software strategies that are completely at odds.

IBM is telling the user community to adopt

OS/2 PM, while Microsoft is advocating DOS

Windows. A great deal of disharmony has been created between IBM and Microsoft.

Yet IBM and Microsoft remain dosely tied. Most

IBM systems are still sold with MS-DOS, which makes Microsoft valuable to IBM, and IBM is still the largest seller of personal computers, which makes it important to Microsoft. Additionally, OS/2 is still vital to both companies as a server platform and even as a future desktop system. But their relationship will never be

This estimate is based on the assumption that

Windows will continue to be a DOS product, one dosely tied to sales of DOS and dependent on sales of Intel architecture PCs. Several recent announcements by Microsoft dearly show that

Windows will move past DOS and even the

Intel PC to other systems and hardware platforms. These announcements are as follows:

• The NT Strategy—An integral part of NT is the indusion of the Windows interface in

OS/2. Thus, Windows and all its applications will run on any platform that OS/2 runs.

• The ACE Initiative—The advanced computing environment (ACE) is a computing system designed for high-performance workstations that use a reduced-instruction-set-computing

(RISC) microprocessor from MIPS Computer

Systems Inc. ACE currendy exists only as a set of specifications from a series of vendors:

Compaq Computer Corporation, Digital Equii>ment Corporation, MIPS, Microsoft, and others.

Microsoft's contribution to ACE is the OS/2

3.0 operating system. OS/2 and UNIX System

V from Tlie Santa Cruz Operation will be the two operating systems that run on ACE computers. Windows will run under OS/2, and consequently will be available in the RISC environment.

With these two aimouncements Microsoft has made a statement: The Windows API is portable. It considers Windows to be a competitive advantage and will implement the Windows API on any platform it deems strategically important.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

CAD/CAM/CAE

Figure 1

Projected and Historical Worldwide Unit Sales of Character-Based and Windows-Based DOS

Millions of Units

16-1

Character-Based DOS

Widows-Based DOS

1989 1990 1991

Souice: Dataquest (July 1991)

Dataquest Perspective

Microsoft has grown to the point that it thinks it can do some things that directly conflia with

IBM's wishes. IBM wants the PM, not Windows, to be the GUI used on every desktop. While

Microsoft is still taking measures to treat IBM well, it is aggressively promoting Windows over the PM and is implementing a strategy of crossplatform availability.

Dataquest believes that the strategy wiD work.

Through the mechanism of OS/2 3-0 (which is mostly written in the C programming language and is itself designed to be portable between

CPU architectures), Microsoft intends to put the

Windows interface on any computing environment that comes into use, whether it be its own or someone else's. In this manner it intends to enlarge its market very quickly.

1992

It is important to realize that by bringing Windows to other platforms Microsoft is not only increasing sales for Windows, it is bringing the entire Windows market to those platforms.

Dataquest estimates that well over 1,200 applications are available for Windows today, double the amount of applications available a year ago.

New Windows products appear every week. So when Windows is implemented on a platform, that platform will receive a flood of sophisticated applications in very short order. Not incidentally, Microsoft is the premier vendor of

Windows applications.

IBM is a co-owner of OS/2. It has the intention to do exacdy the same thing with the PM that

1993 1994 1995

Microsoft is doing with Windows: use OS/2 to make the PM available on multiple platforms.

IBM is at a disadvantage, however, because the

PM application market is small when compared with the Windows market. Given the relative size of the PM and Windows markets, it is dear that Microsoft's strategy is much more likely to succeed.

The long-term implication of the portable Windows strategy is that, as time goes by, operating systems will become less of an issue. Moving between operating systems used to mean writing to different APIs, because each system supported applications only through its own

API. Hence the nature of an operating system and its API was extremely important to software vendors. But with Windows acting as a lingua

franca, the porting of software to multiple platforms will proceed very easily and quickly. The only operating system-specific programming will be that which is necessary to optimize a program for an individual platform. Such optimization, while not an unimportant part of the development process, is a small effort when compared widi that of writing an application for a completely new system.

Bill Gates, Microsoft's chief executive officer, has stated many times that his company's goal is to see a computer on every desk and a

Microsoft produa on every computer. His longterm strategic plans for Windows indicate very dearly that he is not kidding.

By Marshall L. Moseley

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

10 CAD/CAM/CAE

News and Views

Mechanical

AppUcations

Hewlett-Packard to Port

Mechanical-Design Software to Sun SPARCstatlon

Hewlett-Packard Company (HP) and Sun

Microsystems Inc. have signed an agreement whereby HP will port its mechanical computeraided design

(MC:AD)

software to Sun's SPARO station workstations. The initial port will include

HP's ME 10 2-D design and drafting software,

IGES translator, and HP Data Management System. HP expects to make the company's ME 30 solid modeling software products available to

SPARC users at a later date.

Dataquest Perspective

The race to unbundle MCAE/CAD/CAM software is gaining momentum beyond most reasonable expectations. Who would have believed a few years ago that the popular HP mechanical ME

10 product would be offered on a Sun platform.' The market pressure that forced this situation cannot be denied. Large manufacturing companies understand well the need to improve produa reliability and to shorten time-to-market.

Using state-of-the-art CAD/CAM/CAE software with platform independence for many is the most flexible and most cost-effective means of attacking these problems. This growing interest and understanding in the implementation of concurrent engineering does not alone justify the recent flurry of activity by many of the leading software vendors to port to other platforms. The tangible reality of dollars attached to this issue is providing the market pressure to unbundle.

McDonnell Douglas, PDA Engineering, and

Structural Dynamics Research Corporation are all good examples of successful CAE/CAD/CAM companies that owe a significant part of their success to the unbundled approach. When

McDonnell Douglas recently won a large order from General Electric Company with platform independence, the unbundled approach was suggested as one of the compelling reasons for the win. Intergraph Corporation w ^ offer its mechanical EMS product on the Sun platform in the first quarter of 1992. Computervision's rumored alternate platform offering is further evidence of the trend toward unbundling.

The HP software offering on Sun SPARCstation will indude the ME 10 produa, the IGES translator, and HP Data Management System. These tools will give the engineer or designer 2-D design and drafting functions with data management tools. The ME 30 solid modeling product will be ported at some later date. All of these products are supported by HP's Mechanical

Design Division, which functions as an independent software vendor. The in-house direa sales force and VARs will sell both HP- and Sunbased systems.

This is another example of the aggressive marketing stance that HP has taken in the last few months to increase its market awareness and position as a leading vendor in the workstation market.

By Michael J. Seely

GIS

ESRI Goes GUI

A standing-room-only crowd of 2,000 users recently was offered a first look at ARCyiNFO

6.0, a new release presented at the annual user conference for Environmental Systems Research

Institute Inc. (ESRI) of Redlands, California.

Significant product enhancements have been added to the new release, but the addition of industry-standard graphical user interfaces

(GUIs) will have the biggest immediate impact on GIS product requirements. Responding to pressure from the snazzy look and feel of competing products (particularly from archcompetitor

Intergraph Corporation), ESRI's announcement moves the ARC/INFO rich feature set from inscrutable to mainstream. ESRI's aitnouncement finalizes the inevitable: new GIS buyers will now demand, and get, user interfaces that they have seen in trade shows, ads, and computer stores.

Due for release throughout the summer on the many platforms supported, Release 6.0 is nearly three times the size of the previous 5.0 and borders on being a rewrite. The most impressive new feature is the addition of the GRID extensions to ARC/INFO, a set of 200 commands for managing and modeling cell-based raster data. An intertile topology approach provides a method of indexing cross-tile features so that features split by a tile boundary can be treated as one entity. The resuk, virtually continuous maps with invisible cell partitions, greatly facilitates analysis of large continuous

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

CAD/CAM/CAE 11

surfaces. Additional intelligent raster features are promised for future releases. On the vector data side, a new feature labeled dynamic segmentation allows assigiunent of attributes to partial arcs without creating artificial line segments—an obvious benefit to users managing any linear features, such as utility networks.

In addition, previously separate modules such as TIN, Network, COGO, and Database Integrator have been integrated into the core product.

This is, in effea, both a welcome addition in integration and a significant price reduction for buyers that previously would have paid for additional modules. At the same time, significant price reductions were taken on PC ARC/lNFO.

Also announced at the user group meeting was

ArcView, a geographic query-and-display software package for lay users, which was ofifered at $495 for Macintosh and PC versions and at

$995 for UNIX versions. Designed as a productivity tool for those with no GIS skills, ArcView is for users of GIS systems authored by ARC/

INFO users. Described as a "highly leveraged product," it marks ESRI's fijst foray into very low cost software.

Dataquest Perspective

ARC/INFO now does just about everything, on every possible platform, and looks good doing it. Future promised products are right on target: software for analysis of multiple data types and, in particular, the more complete integration of raster and vector tools. In addition, ESRI is doing an artful job of minimizing the presence of INFO, the proprietary RDBMS the company has inherited from Henco Software Inc. of

Waltham, Massachusetts. ESRI's users are young and enthusiastic, and many have the potential of becoming career ARC/INFO proponents. ESRI

President Jack Dangermond, as philosopher and businessman, continues to inspire market growth with his evolving and compelling imagery of an increasingly powerful GIS center serving as architect and author of data visualization tools increasingly in demand from the community at large. Mr. Dangermond's leadership role sells both ARC/INFO software and the

GIS industry itself.

ESRI's next challenge will be to mature with the industry—to evolve from a "research institute" into an enterprise with the financial resources to hold its strong market position.

ESRI management rarely has said "no" to its users, whether the request was adding a new feature or adding a new platform. To support its users 10 to 20 years into the future (not a long time in the GIS mind-set) will require focus and concentration of capital. One of

ESRI's key strengths is its tight integration with users; correspondingly, we believe that one of the hardest things for ESRI to learn will be how to tell those users "no."

By Kathryn Hale

Electronic Design

Automation

Intergraph's Dazix Subsidiary

Signs Technology Pact with AT&T

Inteigraph Ckjrporation subsidiary Dazix recently signed an exclusive six-year technology licensing agreement with AT&T Microelectronics of

Berkeley Heights, New Jersey, to develop and market electronic design systems software tools.

Under terms of the licensing agreement, Dazix will have exclusive worldwide rights to sell, market, and support the design synthesis tools developed by AT&T Bell Laboratories. In addition, Bell Labs and Dazbc will jointly develop high-level design synthesis tools for use in future technologies.

The synthesis software being licensed by AT&T to Dazix was developed by Bell Labs in order to optimize the design of VLSI circuits. The agreement also includes the AT&T Testpilot system, which automates the implementation of scan methodology for application-specific integrated circuits (ASICs).

AT&T showed the synthesis tool set, including

Testpilot, at the Design Automation Conference held recently in San Frandsco. The software tools, much of which are available to EDA users today, with the rest coming by early

1992, address three aspects of design, as follows:

• The AT&T Design Verification System is a mixed-analog digital system simulator that uses the miiVmax ambiguity delay feature recentiy codeveloped by AT&T and Logic

Automation Inc.

• The AT&T High-Level Design System is a performance-driven synthesizer that optimizes

ASIC performance and enables FPGA to gatearray or standard-cell design migration.

• The AT&T Test-Scan System is targeted at automating scan design conversion and

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

12

CAD/CAM/CAE generation. It is complemented by a previously introduced tool, GENTEST, for testveaor generation in combinational and sequential digital ICs and programmable logic arrays.

Dataquest Perspective

Inteigraph's strategy dearly is to use its position and power in the marketplace as a large

CAD/CAM supplier to fill voids in its product line. Intergraph can approach electronics manuliacturers and use them as a large distribution channel. The agreement with AT&T means that Intergraph will be able to strike technology transfer agreements with electronics manufacturers in order to commercialize in-housedeveloped design tools. The advantage for electronics manufacturers is a return on their investment for tools developed; the advantage for companies such as Inteigraph is filling voids in their product line. By being able to get technology fixjm electronics manufacturers and commercialize the technologies, Inteigraph has a solid strategy for providing a technology solution for its customers.

The real challenge for Inteigraph will be to ensure that the technology transfer between the companies goes smoothly. Technology transfers typically are a difficult thing to manage. •

By Ron Collett

In Future Issues

Topics of interest to the CAD/CAM/CAE industry will appear in upcoming issues of Dataquest

Perspective, including the following:

• A look at technical workstation market share

• More "News and Views"

For More Information . . .

Managing Editor Patrick Waurzyniak (408)

437-8132

On the topics in this issue CAD/CAM/CAE (408)

437-8132

On related subjects Client Inquiry Center (408) 437-8671

About other Dataquest publications Sales (408) 437-8000

About upcoming Dataquest conferences Conferences (408) 437-8245

About your subscription Customer Service (408)

437-8402

Via fax request Fax (408)

437-0292

Hie coment of diis report represents our Interpretation and analysis of information generally available to the pubUc or released by re^wnslble individuals in die subject conq»nles, but is not guaranteed as to accuracy or con^leteness. It does not contain material provided to us in confidence by our dients. Individual companies reported on and analyzed by

Dataquest may be dients of this axid/or other Dataquest services. Hiis information Is not furnished in connection 'with a sale or offer «o sell securities or in connection with the solicitation of an offer to buy securities. This firm and its parent and/or their officers, stockholders, or members of their ^unilies may, from time to time, have a loqg or short position in Ihe securities mentioned and may sell or buy such securities.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010869

Dataqyest

a company of

Tne Dun & Brad^rcct Cor poration

Dataquest

Perspective

CAD/CAM/CAE

Vol. 1, No. 1

June 24, 1991

Market Annlysi!*

A View from 30,000 Feet: CAD/CAM/CAE versus the World Computer Market

Dataquest takes a snapshot of the history and the forecast of the worldwide computer industry by platform, comparing this with shipments and revenue in the CKD/CAM/CAE industry over the past 10 years.

By CAD/CAM/CAE industry service staff Page 2

Product Analysis

Schlumberger Jumps on Hewlett-Packard's Workstation Ride

HP's Precision Architecture RISC-based workstations, the latest line of UNIX hot boxes to turn the workstation world on its ear, continue to gain acceptance from the CAD/CIAM/CAE industry, as evidenced by Schlumbeiger's recent adoption of the RISC platform.

By Karen Benson, Laura C. Segeruall, and Michael J. Seely Page 4

Aii.ilvsis

Object-Oriented Databases: A Technology Needing a Kick Start

After years of fragmented research and development efforts marked by slowly evolving product introductions, object-oriented databases (OODB) finally have arrived as a technology. Yet, many questions remain about OODB's commercial market viability. In recent years, however, the advent of object-oriented methodologies, tools, and techniques has combined to spur demand for commercially available OODBs. In this first of a series of articles planned on OODB, this research looks at OODB technology and examines its acceptance by users to date.

By John DeArmon Page 8

EDA Opportunities in the New Design Paradigm

Top-down design and the use of logic synthesis tools are steadily becoming the wave of the future for system designers trying to compete in today's electronics industry. This research analyzes the power of HDL-based design and logic synthesis for the EDA industry.

By Ron Collett Page 12

News <iti(! Views

An analysis of recent news events afFeaing the CAD/CAM/CAE industry.

Computervision Offers Major Upgrade to CADDS Product Line Page 14

VHDL International Consortium Formed by EDA Vendors Page 15

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

001074';

CAD/CAM/CAE

Market Analysis

A View from 30,000

Feet: CAD/CAM/CAE versus the World

Cfnnputer Market

With our first issue of Dataquest Perspective, we thought it would be appropriate to take a few steps back, go up to 30,0()0 feet, and try to position the CAD/CIAM/CAE market in context of the entire computer industry. This overview, which will be followed up with more research on related topics in upcoming issues of Data-

quest Perspective, takes a look at the rest of the computing world, as compared with the CAD/

CAM./CAE world, during the 10-year span from

1986 to 1995.

Major Trends in the Worldwide

Computer Industry

Major growth trends for the general computing industry emerging from Dataquest's research are as follows:

• Single-user workstation unit shipments are increasing by more than a factor of 4

• Supercomputer shipments are forecast to double over the next five years

• PC-DOS unit shipments are expected to double over the next five years

• Every platform segment of the worldwide computer industry is exp)ected to show^ at least modest to significant growth in the next five years

• Host-dependent CPU shipments are forecast to grow at about 20 percent during the next five years

Looking at the computing world from the top,

Figure 1 shows total worldwide factory revenue and total CPIl shipments worldwide on all computer platforms during the 10-year time frame of 1986 to 1995. To provide a perspective of where CAD/CAM/CAE vendors fit into this scheme of things, Figure 2 shows similar history and forecast data for the same 10 years on the

CAD/CAM/CAE industry, of which the revenue totals include that derived from software applications.

In CAD/CL\M/CAE, the primary revenue opportunity is in technical w^orkstations; even though the rest of the computing world will continue to find opportunity for some time in the hostdependent or host-based (mainframe and

Figure 1

Worldwide Computer Industry History and Forecast (1986-1995)

Factory Revenue

180

CPU S h i p m e n t s

60

• 1 Supercomputers

• Host- 40

Dependent

Computers

K l Single- gj,

User ^^

Workstation

PC

Others

[ Z 3 PC DOS

1966

1988 1990

Source: Dataquest (June 1991)

1992

1994 ngss 1988

1990 1992

1994

©1991 Dataquest Incoiporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

CAD/CAM/CAE midrange) systems that have been the staple of big business for many years. As the numbers show, personal computers of all types (including the Macintosh from Apple Computer Inc.) are and will continue to be a substantial percentage of total CPU units shipped.

Most technical workstations currently are going into CAD/CAM/

CAE, where a significant reuenue opportunity exists for vendors of these povuerful machines.

In the general computer market, the PC segment is growing at a 14.3 percent compound annual growth rate (CAGR). Much of this growth can be traced to the diversity of vertical market applications that are widely used across the board. Historically, each application has been used in a particular type of industry or job category. More sophisticated applications and combined applications are the impetus for growth in the future. The advances in computing technology are moving into each platform type, resulting in price/performance benefits for all. The PC community has been particularly aggressive in this regard.

This general platform growth profile is mirrored in the CAD/<>M/CAE industry. Similar to the general industry, single-user workstations have the highest growth rate. Single-user workstations shipments will grow by a factor of 4 over the next five years in the total market. However, single-user workstations will grow by only a factor of 2.5 times in the CAD/CAM/CAE industry. The difference between the faster growth in technical workstations in the overall computing world versus CAD/CAM/CAE is due to the growth of other technical applications, such as computer-aided software engineering (CASE), and also to the inroads of technical workstations into nontechnical applications, such as those supporting financial securities banking.

In the PC arena, although the PC industry is growing by a factor of 2 in terms of CPU unit shipments, CAD-related applications are growing at about 30 percent. The current PC industry trend toward portable and, in the very near future, pen-based PCs is generally given credit for fueling the growth in the PC market. Steadily improving price/performance throughout the

PC industry is making powerful machines affordable to all. In the future, these trends will reflect onto the CAD/CAM/CAE industry. In faa,

Dataquest expects a small CAD market for portables to emerge over the next few years.

Figure 2

Worldwide CAD/CAM/CAE Industry History and Forecast (1986-1995)

Factory Revenue

25m Host/

Server

CPU S h i p m e n t s

1200-

Host/

Server

Workstation ^ d ^ ^ t'^j^'l PersonoU j^^^^

I I Technical

Workstation

20-

15600

10-

400

5-

0-

rriTI^^^^^^i^

1986 1988 1990 1992

1994

200

Source: Dataquest (June 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

CAD/CAM/CAE

Although a large installed base of host-based

CAD produas exists, the growth of the CAD/

CAM/C:AE

market is based on technical workstations and personal computers. Dataquest expeas the host-based (lAD/CAM/CAE market to experience a modest downturn as a result.

In 1S>91, CAD vendors shipped approximately

593,980 CPU units compared with approximately

27.9 million CPU shipments worldwide. This apparendy small share of the total CPU market is distorted by the large share held by PC-based products. Of the total worldwide CPU shipments, 26.3 million were personal computers.

CAD/CAM/CAE holds only a minuscule share of the total PC market, accounting for only

1.7 percent of the total shipments of PCs worldwide. This 1.7 percent does represent the high end of the PC market in graphics and computing performance. At the other end of the spectrum, CAD is a major force in the technical workstation area, and it is an important application on high-end PC products. CAD/

CAM/CAE is a very significant driver for the technical workstation market, representing

35 percent of the total CPU shipments worldwide.

Dataquest Perspective

Most technical workstations currentiy are going into CAD/CAM/CAE, where a significant revenue opportunity exists for vendors of these powerful machines. Although workstation makers are trying to infiltrate many other types of commercial applications, it is in technical applications, such as

CAD/C:AM/CAE,

where technical workstations will continue to find their stronghold.

Although the CAD world is focused on workstations, the rest of the computing world will continue to generate significant revenue, from PCs to supercomputers. For example, factory revenue for hardware and software in CAD/CAM/

CAE reached the $14 billion level in 1990, while the entire computing industry during the same year had total factory revenue (hardware shipments) of $119 billion. As users demand more and more heterogeneous computing environments, computer vendors must learn to adapt to the resulting real-world complexities.

In the next issue of Dataquest Perspective, the

CAD/CAM/CAE industry service will present an overview forecast of market and price trends of scalar-compute power. This research will analyze the various modes of determining compute power and how the measure of millions of instructions per second (mips) varies over the range of hardware from supercomputers to personal computers. •

By CAD/CAM/CAE industry service staff

Product Analysis

Schlumberger Jumps on Hewlett-Packard's

Workstation Ride

(Note: Some of this research is reprinted with the permission of the Technical Computing System industry service and European Computer industry services, uHth additional material provided by the CAD/CAM/CAE staff.)

Exdted by the stellar performance numbers staked out by latest Precision Architecture RISCbased workstations from the Hewlett-Packard

Company (HP), Schlumberger Technologies recendy became the latest CAD/CAM/CAE software vendor to hitch its wagon to the lightning-fast HP UNIX workstations introduced in March of this year.

Witii the HP Apollo 9000 Series 700 and 400 workstations, HP recaptured the lead in highperformance workstations and seized the speedmerchant bragging rights from Sun Microsystems

Inc., IBM Corp>oration, and Digital Equipment

Corporation, as the new HP boxes greatiy exceeded the price/performance benchmarks of those vendors' technical workstations. As a result, HP and Schlumberger teamed up last month to ink an agreement valued at $30 million, under which Schlumberger (Ann Arbor,

Michigan), the subsidiary of Schlumbeiger Ltd., will resell the HP Apollo 9000 Series 700 and

Series 400 workstations bundled with Schlumbeiiger's DesktopBravo! line of CAD/CAM software.

The introduction of the RISC-based workstations enabled HP to streak into the lead in the workstation price/performance stakes, taking a priceleadership position with a dear "Try matching that!" ultimatum to Sun Microsystems, HP's main sparring partner in the workstation market. With the Series 700, HP truly has set a new performance level for desktop computing (see

Figure 1).

With performance quoted at 55.5 SPECmarks,

HP's new systems offer performance that is more than 1.5 times that of their closest competitor, IBM's RS/6000 Model 320H. Not only did HP push performance, it also aggressively priced the systems. In brief, the three models in the Series 700 offer from 57 to 76 mips and

17 to 22 mflops, within a price range of

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

CAD/CAM/CAE

Figure 1

Comparison of Competitive Systems

5^S^5^^^

HP 900

Series 720

Digital DECstatlon 5000

Sun Microsystems

SPARCstatlon 2

^s^^s^^

IBM RS/6000

Model 320H

10

^WS^'^^F:^^^

"^rw:

20

30 40

SPECmarks

SPEC Integer

•Vd SPEC Floating Point

50

60 70

BO

Source: Vendor literature, Dataquest (June 1991)

$11,990 to $43,190, leaving IBM and Sun behind with the price/performance curve of their corresponding technical workstations, the

RS/6000 and the SPARCstation 2, respectively.

Second-Generation Precision

Architecture RISC

The new^ family is based on HP's proprietary second-generation CMOS Precision Architecture

(PA)-RISC architecture, extending the company's

PA-RISC offerings from the Series 800 family of business systems and servers. In its product line spanning commercial scientific, technical, and engineering computing environments, HP has made a full commitment to RISC technology and open systems, offering modularity, scalability, and the all-important investment protection.

The second-generation PA-RISC architecture incorporates several features of HP's high-end

PRISM workstation architecture—technologies that were attained through the acquisition of

Apollo. For example, the PA-RISC architecture and the Series 700's graphics operations are tightly integrated, much in the same way as the graphics subsystem in the PRISM-based Apollo

Series 10000 is integrated with that system's

RISC CPU.

The family comprises three models—the 720,

730, and 750—available with a variety of graphics options from gray scale to very sophisticated

3-D modeling. All models run HP-UX, HP's proprietary implementation of UNIX and OSF/1.

Both the Models 720 and 730 can be configured with memory of up to 64MB, 128K instruction, and 256K data cache; and internal disk capacity of up to 840MB (2 x 420MB) and external of up to 10GB maximum (7 x 1.3GB external with plug-in SCSI-II). One EISA slot is an option on the Model 720 and comes as stan<^rd on the Model 730. Both can be configured as desk-side or desktop systems.

The Model 750 provides the same performance as the Model 730 but offers expandability to operate as a large server plus kiger cache

( 2 5 6 K ) .

The desk-side Model 750 can support up to 192MB random-access memory (RAM) and has an internal storage capacity of 2.6GB.

This model also offers a standard graphics connect slot and four EISA slots. The Model 750 also supports internal, removable mass storage devices, such as CO-ROM, 4mm digital audio tape, and floppy disk. The operating system, applications, documentation, and updates can be supplied in CD format that, according to HP, will lower users' installation, support, and

©1991 Dataquest Incorporated / 1290 Bidder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

CAD/CAM/CAE documentation costs. The standard package comes "With integrated 802.3 Ethernet Support and support for all HP-UX 8.0 networking capabilities, FDDI, TCP/IP, NFS, and LAN Manager.

Operating Systems

Similar to the Series 800 of multiuser systems, the Series 700 supports the HP-UX operating system, which is based on and is compatible with USL's UNIX System and is also POSDCand XPG3-compliant. In the long term, however, HP's strategy for UNIX operating systems revolves around the Open Software Foundation's

OSF/1. OSF/1 will be available on the Series

700 family in the second half of 1991; in fact, the company is labeling the new products as

"HP's first OSF/1 platform." HP's intention is to offer OSF/1 on its business servers, when the commercial OSF/1 version of UNIX matches the functionality of HP-UX.

Applications and Target Markets

Hewlett-Packard counts more than 2,000 applications available on PA-RISC today, most of which stem from- the Series 800 minicomputers and mainframes. SoftPC also gives access to more than 50,000 MS-DOS applications. On the technical side, HP has deployed Series 700 models to more than 100 independent software vendors (ISVs) for qualification, aiming to encourage software development in such areas as simulation, mechanical design, molecular modeling, satellite management, and flight patterns. Success is obvious because all major multiplatform vendors in the CAD/CAM/CAE industry have been choosing to use the new HP workstations over competing workstations in a wide variety of applications demonstrations.

Graphics

A fiill range of graphics options is available, providing windowing performance with the X

Window System, OSF/Motif, and HP VUE. The graphics capabilities begin at the lower-cost

GHW gray-scale option (910,000 XH vectors/ second), through the CRX, 2-D, and 3-D color wireframe at 910,000 Xll vectors/second, and more than 1 million 2-D and 3-D vectors/second. The TVRX—^TurboVRX high-performance

3-D modeling and rendering graphics system— provides high-performance anti-aliasing, alphablending, and texture-mapping capabilities.

Dataquest Perspective

Hewlett-Packard has resoundingly answered industry and customer criticism about its lack of a low-cost RISC workstation with these high-performing systems. Perhaps the only concern that now remains about the company's workstation strategy is its multiarchitecture approach. The company's •workstation family now looks as shown in Figure 2.

The PRISM-based DNIOK products remain at the high end of the scale, despite the fact that, as described earlier, some features of this architecture have been incorporated into the secondgeneration PA-RISC. The fact that HP has started to integrate PRISM-like features into PA-RISC would indicate that in the third-generation PA-

RISC, which is expected to take place within the next two years, more than full integration of PRISM technology may drive CPU performance to more than 100 mips. In fact, HP already has PA-RISC prototypes running at more than 200 mips in its laboratories in Boeblingen,

Germany. Another benefit of this development strategy would be the reduction of concurrent architectures to two: Motorola and Precision

Architecture.

Hewlett-Packard insists on the company's loyalty to the Series 400 Motorola 680X0-based family of workstations and will reinforce this loyalty with upgrades later this year or early in 1992.

The company has signed a large order with the

Singapore stock exchange for delivery of Series

400 workstations for financial analysis. With a large installed base to placate and a wide range of commercial and tedinical applications to oiTer, the company is likely to continue upgrading the Motorola line until well into 1992 at least. However, following the company's claims regarding the tow cost of its CMOS RISC technology, a low-end RISC offer may be expected, aimed at the commercial market and competing with the Sun SPARC IPC workstation in 1992. If these machines match the price/performance ratio of the Series 700, in the longer term, the

Motorola family could fade into the background.

Given the current trend of margin erosion and price cutting in the workstation market, HP is wise to set its sights on an homogeneous line,

Digital Equipment is certainly "feeling the pinch" with regard to supporting its twopronged workstation architecture and operating systems approach, with VM S/CISC and

ULTRIX/RISC MlPS-based workstations. Hewlett-

Packard is resolving the architeaure and also the operating-systems issue with the phasing out of Apollo's Domain operating system and the expected implementation of OSF/1 on its workstations. The company predicts that 90 percent of the Series 700 will be running OSF/1 by die end of 1992. Where this leaves HP-UX in the long term remains to be seen.

Having lost market share in 1990, HP now has all the necessary assets not only to recover but

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

CAD/CAM/CAE

Figure 2

Hierarchy of HP's Workstation Line

Price

i

i

r

y

X Terminals

DN10K

PRISM

Series 4D0

6dK

PA-RISC

Performance

/

Applications

Availability

Source: The Hewlett-Packard Company to poach other vendors' potential sales. However, it is imperative that availability of these models is not delayed, as other RISC players also have plans for higher-performance architecture. MIPS Computer Systems (Sunnyvale,

California) has already revealed its development of the R4000 64-bit processors planned for

1992, and it is unlikely that Sun Microsystems has been sitting back, arms folded. At present, though. Sun appears to be busy targeting the high-volume market, competing with high-end

PC vendors such as Apple Computer Inc. and

Compaq Computer Corporation.

Hewlett-Packard claims to have harvested enthusiastic responses from ISVs with prelaunch implementations of the Series 700 such as,

"Up and running in one day, no recompile— performance outstanding," courtesy of a leading database ISV; and, "Fastest workstation to date," from a leading GIS vendor. With such impressive performance figures on report, such claims may indeed be justified.

Ifie Schlumberger/HP Connection

All of the major mechanical CAD/CAM/CAE software vendors ofifer software products on one or more of the big four hardware vendor platforms. A closer analysis indicates why

Table 1

Leading Mechanical CAD/CAM/CAE Vendors;

Worldwide 1990 Software Revenue o n

Technical Workstations

(Millions of U.S. Dollars)

Company

C^mputervision

McDonnell Douglas

SDRC

Schlumberger

Hewlett-Packard

Source: Dataquest (Jime 1991)

Revenue

165.5

65.2

61.1

46.0

39.9

Schlumberger jumped on the HP Series 700 bandwagon. Last year, Schlumberger ranked fourth in market share for software revenue on technical workstations (see Table 1) behind

Computervision, McDonnell Douglas, and Structural Dynamics Research Corporation (SDRC).

The incremental business opportunity for

Schlumberger with HP, added to the company's already significant DEC-based product sales, could vault the company forward one or two steps to second place in market share, right behind Computervision.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

8

CAD/CAM/CAE

Although Schlumberger's strategy takes advantage of the significant price/performance of the new product offering fk'om HP, one cannot fail to recognize the fact that McDonnell Douglas and SDRC have been using HP products for some time. The value of installed base and a trained sales and support staff will dilute some of the incremental business potential for

Schlumbeiger.

As the level of software reaches ever higher levels of sophistication, hot technical workstations are indispensable in maintaining reasonable performance levels. Many software vendors should thank HP for extending the useful life of millions of lines of FORTRAN and C code. If this trend continues, as Dataquest expects it will, massive amounts of software •will be in a better position to be enhanced with new user interfaces and embedded processes for data translation or other tasks that automate the user's job. Perhaps more important, this trend will make it easier for the software developer to maintain upward compatibility when upgrading complex software packages. •

By Karen Benson

Laura C. Segervall

Michael J. Seely

Technolu;4y Analysis

Object-Oriented Databases: A Technology

Needing a Kick Start

The objea-oriented database (OODB) market has finally arrived. After nearly a decade of fragmented proprietary research and development efforts and product offerings from hardware and software vendors, nonproprietary

OODB vendors have introduced productionquality products. Supported by the advent and adoption of object-oriented methodologies, techniques, and tools, the demand for commercially available OODBs has increased significantly over the past decade.

At present, the OODB market is still embryonic, with total 1990 annual revenue for OODB vendors reaching approximately $9 million. The most significant players in the OODB market include Object Design Inc. and ONTOS Inc.

(formerly (Dntologic Inc.), both based in

Burlington, Massachusetts; Servio Corporation of

Alameda, California; and Objectivity and Versant

Object Technology, both of Menio Park, California. Most market opportunities in the near-term time frame will be primarily through the OEM and VAR channels and, more spedfically, with third-party applications and tools vendors, hardware vendors, and vertical telecommunications vendors.

The vast majority of users today simply do not have the knowledge, capability, or willingness to implement and maintain a commercial OODB-based system successfully. This technology-transfer factor will remain a crucial element in the development of the OODB market. Over the next five years, the growth of the commercial OODB market will depend upon several critical factors including the following:

• The integration of commercial OODBs into third-party vertical applications and tools

• Overall adoption of object-oriented technology in the marketplace

• Tools availability for implementation of

OODB

• The proliferation of distributed, interoperable computing environments

Hardware vendors' make versus buy decisions

• Technical applications and tool users demand for commercial OODBs

Two plausible near-term (five-year) growth scenarios may result from these critical faaors:

The market either will undergo a moderate rise, or it will experience a robust growth in market demand for commercial OODBs. These two scenarios and related critical factors to the market will be discussed here in general as well as in subsequent articles. Future issues of

Dataquest Perspective will include coverage of the major OODB vendors, market dynamics, end-user research, and other related topics.

Prolonged Sluggishness for OODB

The first potential market scenario is a continuation of the current sluggishness in this market.

This inactivity is primarily the result of the three following significant factors:

• The reluaance and inability of ISVs and hardware vendors to usefully incorporate commercial OODBs in their product offerings

• A lack of aftermarket tools and services for

OODB vendors to sell back into the OEM and VAR channel

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-SOOO / Fax (408) 437-0292

0010745

CAD/CAM/CAE

• The relatively high learning curve faced by end users in the migration, creation, and maintenance of computer systems that utilize

OODBs

If the aforementioned factors inhibiting OODB growth remain unchanged, the result will be a

59 percent CAGR over the next five years in an

OODB market that will amount to approximately $92 million in 1995 (see Figure 1). However, if OODB vendors are able to address these issues successfully, Dataquest expects the

OODB market to grow at a much more robust pace of 104 percent CAGR, reaching about

$320 million in revenue by 1995. The conditions that would create this second, more optimistic, scenario are discussed as follows.

Although a few agreements have been reached between each of the major OODB vendors and

ISVs and hardware vendors, most of the partnership relationships have been joint developments, partial technology transfers, and other arrangements that, to date, have not resulted in significant revenue streams for the companies involved. Hardware vendors and ISVs have had difficulty incorporating commercial OODBs in a fashion that adds p>erformance and functionality while also allowing for customization and migration of existing user data and applications.

As users customize their data models and systems, the attempt to migrate to a commercial

OODB becomes a daunting task, at the least.

OODB's Target Vertical Markets

Historically, the primary targets of OODB vendors have been the CAD/CAM/CAE and CASE vendors in the technical computing market—^in addition to vendors in on-line transaction processing (OLTP), document image management systems (DIMS), and electronic publishing market analysis (EPMA) in the general-business

DP/MIS markets. These tai;get vertical markets will remain stagnated for OODB until vendors begin to understand the differences between available OODB implementations and also how these OODBs can be utilized successfully in their tools and applications.

Objea-oriented systems can mean many things to many people. Currendy, a large number of different object-based systems are installed in these vertical markets, creating a difficult integration and evaluation challenge for the

OODB vendors and the ISVs and hardware vendors. In a recent Dataquest survey of more than 100 large end users of CAD/CIAM/CAE and

CASE/Software Development Tools (SDT) products, 39 percent of users have incorporated standard data models at their companies. These standard data models, however, are not consistent among users. Moreover, only 6 percent of the 6l percent of users who have not yet incorporated standard data models intend to do so within the next year, and less than 1 percent intend to do so during the next two years.

A major challenge for the OODB vendors will be to address the issue of the lack of standard data models and the apparently slow movement toward such, as well as the existence of disparate data models between and within various applications.

End-User Reluctance

The main end-user factors affecting OODB market development include the lack of robust

OODB maintenance and development tools and services, as well as the lack of knowledgeable and uniform object-oriented technology understanding. Most of the vendors have developed, or are developing, tools for their OODBs. The problem, however, is the faa that a fool with a tool is still a fool. Providing tools to unknowledgeable users can be self-defeating, in that typically the tools will be perceived as not being able to deliver as promised. This lack of knowledge has also been a well-documented problem in the CASE and CAE markets. Moreover, the modest 39 percent of users incorporating standard data models is a direct reflection of the level of sophistication of the end-user population—only the most sophisticated end users incorporate standard data models. Therefore, OODB vendors must provide extensible and scalable tools for OODB implementation and maintenance, as well as services for education and training.

The low level of user understanding of objectoriented technology also has caused a noticeable reluctance by end users to pursue OODB implementations at their sites. Based on its enduser survey, Dataquest found that 32 percent of end users believed that OODB was a highly important part of their strategic plan, 32 percent deemed OODB to be moderately important, while the remaining 36 percent viewed OODB as having a low level of importance (see

Figure 2).

These data suggest that, although there is a clear interest in OODB as an enabling technology and competitive advantage in the CAD/

CAM/CAE and CASE/SDT markets, a significant number of users still have not decided if they are committed to OODBs, and an even laiger

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

10

CAD/CAM/CAE number of users are uninterested in OODBs.

Moreover, closer examination of the survey data reveals a polarity in the distribution of importance of OODB across these users, with 25 percent viewing OODB as very low in priority and

17 percent deeming OODB as very high in priority. This polarity suggests that users have decided already, or are in the process of deciding on OODB use within their organizations.

F ^ r e 1

Forecast Growth of Commercial OODB Market

Millions of Dollars

325-

285-

245-

205-

165-

125-

85-

45-

1

1989 1990 1991

Source: Dataquest (June 1991)

1

1992

Critical to the success of OODBs will be the movement of these "fence riders"—the majority of the users who value OODB as medium in importance—either toward or away from OODB use.

Of the users who rated OODB in the highest category of importance, 41 percent have a standard data model of some type (see Figure 3).

Robust Growth /

(Scsnarto 2) /

\ /

\y

j f ^ Modest Growth

j ^ (Scenario 1)

[ 1

1993 1994 1995

K i

Fi^;ure 2

Perceived Level of OODB Importance

10

5

0

Percent of CAD/CAM/CAE and CASE/SDT Users

40p.. V \ V W v \ "•-> "K 35

30

25

20 ^

15

VFSSSSSSSS\

vixww^.

^^sm

v.\/v

^^:m

^i i

High

.%.v

Low

Source: Dataquest (June 1991)

©1991 Dataquest Incorporated / 1290 Rldder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

i

CAD/CAM/CAE 11

This is good news for OODB vendors because more than one-third of users who value

OODBs have systems in place that can be migrated more easily into a commercial OODB.

Furthermore, the users who rated OODB at either low or medium importance have a standard data model in place, which eases possible adoptability reluctance. The real challenge for

OODB vendors lies in dealing with the issues of object-oriented applications and a willingness to implement these solutions.

Perhaps the most promising note for OODB vendors is that of the users who rated OODB as being high in importance, 54 percent have adopted, or plan to adopt, a commercial database within the next year (see Figure 4).

Although this fact does not affect commercial

OODBs directly, it clearly shows that the interest level in commercial databases is high among those who believe that OODB is important. Among those users who rated OODB as

Figure 3

Use of Standard Data Models among OODB-Interested Users

Percent of Users Who Implemented Standardized Models

60-

50

40

30-j

20

"-'', "\. "\ "^.W'vV'O

^»>C<"<-X'x^-

<\V^XVw%?

?$$$j$m

'd^X^s^'K^

'r. V \ t : 4> ->

VP

-'V •']

'*1-^ "''- ' ' \ " t ''''/ \ "<(r.- ^^ • \ .

10

^ ^

Low Medium High

Perceived Level of Importance of OODB by CAD/CAM/CAE and CASE/SDT End Users

Source: Dataquest (June 1991)

Figure 4

Use of Commercial Databases among OODB-Interested Users

Percent of Users Intending to Use a Commercial Database within the Next Year

60-

50

40

30-1 iZVXVXXXXW

^mmi

'\\\ '•:. \. \ \. "•', \

20

10 ^ ^ ^ ^

. ^

Low

I^^^^NXSKSN

Medium

Perceived Level of Importance of OODB by CAD/CAM/CAE and CASE/SDT End Users

' ^ . ^

R S ^ S ^

High

Source: Dataquest (June 1991)

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

12 CAD/CAM/CAE low or medium in importance, 34 percent and

39 percent, respectively, have adopted or plan to adopt a commercial database within the next year. Whether these users adopt a commercial

OODB rather than another type of database is questionable, but it is of immediate importance for OODB vendors to quickly provide easy-touse tools and services that w^iU allow these users to adopt OODBs as a solution.

Dataquest Perspective

Object-oriented methodologies and technologies hold great promise for the future of the electronics marketplace. The role of OODB in the market has yet to be determined, but the window of opportunity for the OODB vendors has opened. Just how long this window remains open, and how OODB vendors, ISVs, and the hardware vendors decide to take advantage of the OODB opportunity, has yet to be determined—but these are the critical questions.

Dataquest believes that if OODB vendors provide the right tools and services to ease the migration, creation, and maintenance of OODBbased systems as well as address the question of the technology-transfer problem facing end users, the probability of robust growth is very good. All of these items are critical to OODB vendors in that they will lower the cost for adoption to end users and further the likelihood of OODB acceptance.

Further, Dataquest also believes that if the

OODB vendors can integrate with strategic

OEMs and VARs in selected vertical niche markets where the vendors have expertise, the chances of realizing the robust growth path discussed earlier are even greater. Because providing tools and services to end users is a very expensive—and often unprofitable—business, it will be of the greatest importance for the

OODB vendors to specialize in strategic vertical arenas and, thus, effectively manage this part of their operations.

By John DeArmon

EDA Opportunities in the New Design

Paradigm

As the cost of silicon continues spiraling downward, the price per gate on a gate array, or cell-based IC, is no longer a significant barrier to implementing high-gate-count ASICs. In 1990, for example, the cost per gate of a CMOS gate array ranged from 40 millicents to 120 millicents, with the average hovering around 80 millicents. Since 1983, the price has decreased at a

CAGR of 32 percent. Undoubtedly, this era of low-cost silicon will help pave the way for higher-gate-count ASICs. But low-cost silicon alone will not usher in the era of high-gatecount ASICs.

Indeed, with market windows and product life cycles constantly shrinking, designing ASICs in the 250,000-gate to 500,000-gate range (in say 9 to 12 months)—or even 1 million gates for that matter—will be overwhelming, if not impossible, for most companies. In other words, it does not logically follow that low-cost silicon translates into higher-gate-count ASICs. The real issue with respect to low-cost silicon is whether designers will be able to exploit the silicon in the face of shorter and shorter design cycles.

Dataquest's research shows that top-down design, which includes the use of a hardware description language and a logic synthesis system, is certainly helping the plight of today's designer. Figure 1 illustrates the power of the tools when put in the hands of the most skilled engineers. In the mid-1980s, a design engineer using a schematic-entry system and gate-level simulator could design approximately

700 gates per month. In the first-generation, preproduction release, top-down design tool environment, the designers' productivity increased more than threefold, to approximately

2,500 gates. And when finally equipped with a production version of the tools, the designers' productivity increased to 4,500 gates. Thus, a leading-edge designer can design approximately

50,0(X) gates per year.

Today's leading-edge merchant tools are capable of supporting 100,000-gate to perhaps

2(X),000-gate designs. But how will companies design tomon-ow's 300,000- to 500,000-gate devices? Undoubtedly, the tools and workstations will improve, and this will help the situation. But Dataquest believes that the advances made in silicon processing over the next three to five years will continue to outpace the advances made in EDA software and workstation performance. In short, despite the fact that workstation perfonnance will quadruple by 1995 and EDA software will reap the benefits of added compute pxjwer, the EDA software tools will stiU lag the advances made in silicon processing. For example, 0,5-micron fabrication processes are expected to be in place by the

1995 to 1996 time frame. So what's the solution?

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

CAD/CAM/CAE

13

FJ^re 1

EDA Productivity Analysis

Gates/Month/Engineer (Design Engineers)

8000-

1

import of HDC-Based Design Reusabiilty

7200

6400

5600

4800-

1 -^. ^

(jurrent-ueneratio n yyntnesis

^ ^

\

First-Generation Synthesis (Beta)

\

\

4000

3200

2400

1600

Gate-Level Design

1

'

600

1985

A9^

Source: Dataquest (June 1991)

198S 1991

yi!m>

The Solution: A New Form of

Design Reusability

Dataquest believes that there is only one clear solution: Design teams must reuse large portions of previous-generation designs. Reuse has been a topic of discussion for many years, but only a modest amount of reuse actually takes place. Dataquest estimates that only 5 percent to 10 percent of an average ASIC design uses logic from a previous design. This, of course, does not include the reuse of standard macrocells found in an ASIC library, although the use of such a library is in fact reuse. The kind of reuse we are concerned -with is the functional blocks comprising these standard macrocells.

There are several reasons for the lack of design reuse. For instance, comprehending a high-gatecount logic block represented as a schematic can be very difficult if the block's capabilities are neither characterized well nor documented clearly. Also, engineers have a tendency to want to improve on a previous design and, thus, attempt to reengineer a design, as well refrain from using another designer's work.

In order for reuse to grow, a new reuse paradigm must emerge over the next five years that will facilitate reuse. Dataquest believes that this new paradigm will emerge and that it will take the shape of reusable logic blocks represented in hardware description language (HDL) form that are resynthesizable. In other words, ASIC designers will describe logic functions in HDL, and the HDL functions then will be archived using a sophisticated library management system. Designers will retrieve these functions and reuse them on subsequent designs.

These fiinaions might include anything from a custom-barrel shifter to perhaps a SCSI controller. Electronic systems manufacturers that are most effective in implementing an efficient

HDL-based reuse methodology and environment will be the winners in the electronic systems markets in which they compete.

In Dataquest's view, HDL-based design reuse offers two key advantages over schematic-based design reuse. First, a design represented in HDL design is inherently easier to comprehend than a schematic design. A design in HDL is textual and can be commented. A good analogy is to compare the task of comprehending a software program written in a high-level language (such as C) with the same program written in assembly language. Second, HDL-based designs

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

14

CAD/CAM/CAE provide a much greater degree of foundryindependence than a design that is represented in schematic form only. For instance, an archived HDL block can be resynthesized to target a different ASIC vendor's macrocell library.

Dataquest believes chat design reuse will grow dramatically over the next five years such that by 1996 only 30 percent to 50 percent of a given ultra-Iarge~scale integration (VLSI) design

(approximately 400,000 gates) will comprise original design work. The remainder will be taken from existing HDL designs, as well as hardwired megacells developed by the ASIC foundry.

Certainly, hardwired megacells offered by an

ASIC vendor might seem to obviate the need for equivalent HDL-based megafunctions, but such hardwired functions have several inherent drawbacks that will cause many designers to turn to an HDL-based reuse solution. First, the hardwired function is not foundry-independent, and foundry independence is something that the market wants. (Dataquest's research shows that the ASIC market prefers to work with two to three different ASIC foundries to ensure turnaround time and volume produaion.)

Second, before an ASIC manufaaurer will commit to offering a hardwired megacell, the ASIC manufacturer must be sure that the development justifies the return on investment. Dataquest believes that, in many cases, ASIC vendors wiU opt out of offering certain megacells because the return on investment is outweighed by the cost of development and support.

Furthermore, many ASIC vendors with standard

IC produa offerings will be reluctant to offer an equivalent megacell version of the standard

IC for fear that such an offering will cannibalize existing (and perhaps lucrative) standard product offerings. These vendors will offer such an equivalent version only when it becomes clear that they have no other choice—offer this version or lose the business to a competitor supplying the cell in its ASIC library. Finally, in many instances the ASIC vendor will be unable to offer a hardwired macrocell in the time frame demanded by the market, and so the

ASIC market will have no choice but to develop it in-house (assuming, of course, that use of an equivalent standard IC has been rejected).

Dataquest Perspective

The Opportunity for EDA Vendors

As HDL-based design reuse moves into the mainstream, the result will be a significant increase in ASIC-device gate densities. For EDA vendors, the market opportunity will be to provide sophisticated design automation environments that can support HDL-based reuse. These design environments include not only advanced application software capable of handling highdensity ASICs but also data management technology.

For example, Dataquest believes that library and design data management will become critical to achieve maximum reusability of HDL blocks.

The system will have to keep track of a variety of files and application tools to ensure that the block can be resynthesized successfully at a later date. For instance, all of the simulation and timing verification files must be linked to the HDL block file, the respective versions of all of the application software packages used to design and synthesize the block must be maintained, and the version of the ASIC library must be stored. •

By Ron Collett

News and Views

Computervisicm Offers

Major Upgrade to

CADDS Fr€}duct Line

Computervision, a subsidiary of Prime Computer

Inc., introduced a major revision of CADDS 5, the company's CAD/CAM software, in an update that includes many significant enhancements in user productivity plus improvements in speeding up the product-<iesign cycle.

CADDS 5, which will be provided to most current users of CADDS 4 at no additional charge, incorporates a host of advanced technologies such as parametric modeling, variational geometry, a sketcher, an intelligent user interface, a constraint modeler, feature-based modeling, and

X Window access along with an open database architecture and a comprehensive suite of integrated applications.

Computervision's CADDS 5 gives engineers the first production implementation of parametric modeling and variational geometry that is integrated with proven applications for design and detailing, engineering and analysis, and manufacturing. The new modeling environment of CADDS 5 allows engineers to capture design

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

CAD/CAM/CAE

15 intent and quickly make changes to a CAD part or assembly.

Dataquest Perspective

The new CAD/C:AM/CAE architecture developed for CADDS 5 provides a staggering array of geometric modeling and user interface improvements in addition to enhancing access to an already extensive list of Computervision application software. A shopping list of every hot new geometry-related technology in recent years has been included in this release. What could potentially be a mishmash technology du Jour has been crafted into a well-orchestrated complement of geometric construction techniques.

The integrated approach allows the designer to mix and match constraints, sketching, and features with surface- and solid-modeling technology. This gives the designer full flexibility to roam in the previously troublesome waters of simultaneous detail design and conceptual design.

The enhancements to the user interface has transformed what was a previously tired and outdated CADDS- 4 command- and menu-driven user interface into the state-of-the-art realm by using adaptive defaults, icons, and pop-up and fanfold menus with flexible custom features.

Two standard interfaces are shipped with the product, one for the power user and the other for the occasional user—the result is very friendly.

Easily overlooked in the excitement is the single, compatible product database. The benefit this database brings to the individual is difficult to measure. Some of our survey data suggest that the overhead associated with the nonproductive activity of making disparate databases interact can reduce the effectiveness of the designer by 30 percent or more. Errors in conversion and lost data are the most common ailments. A single, compatible product database alleviates this frequent problem. CADDS 5 can solve this problem for a wide variety of workers from concept designers to purchasing agents. TTie value of a common database is compounded every time one worker passes engineering data to the next person in the process.

CADDS 5 is the closest thing to a framework that the mechanical CAD/CAM/CAE users are likely to see in the next few years. Moving this technology into the 150,000-seat installed base may set the stage for market leadership in this area.

Technology is only part of the picture. Without a sound pricing and packaging scheme, many great technical achievements have dried up without a hint of market success. The CADDS 5 introduction could act as a case study for how it should be done. Apparendy, the marketing people in Computervision actually talked with some users before setting the pricing and packaging strategy. If you want to encourage users to upgrade to the new release this year, what is a good fee? Zero dollars. If you want the users of old CDS- and Motorola-based hardware to upgrade to the new release software, tell them to buy a SPARC-based workstation and the software fee is zero dollars. Many of the application packages have been reduced in price and combined in packages for the power users and specialists.

The next CAD/CAM/CAE issue of Dataquest

Perspective will include a more detailed analysis of CADDS 5 after a hands-on evaluation is completed, and more information on new pricing will be outlined as well.

By Michael J. Seely

VHDL IntematUmal

Consortium Fortified by

EDA Vendors

In an aaempt to establish the VHSIC Hardware

Description Language (VHDL) as the standard for electronic design automation, several EDA and ASIC vendors, along with VHDL users, recently formed an industry consortium called

VHDL International Inc. VHDL International's founding members include COMPASS Design

Automation, LSI Logic Corporation, Logic Automation Inc., Mentor Graphics Corporation, Synopsys Inc., VLSI Technology, Valid Logic Systems, Vantage Analysis Systems Inc., and

Viewlogic Systems.

In addition to the support of the aforementioned EDA and ASIC vendors, the VHDL Users

Group (VUG) endorsed VHDL International, forming a subcommittee of the group that will play an important role in helping to define and recommend solutions to issues on hardware description languages.

VHDL has encountered oiganized opposition since Cadence Design Systems Inc. (Sunnyvale,

California) opened up its Verilog Hardware

Description Language (HDL) last year and established a similar consortium, Open Verilog International (OVI). VHDL, defined by a user

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

16 CAD/CAM/CAE committee in 1981 and ratified by the Institute of Electrical and Electronics Engineers (IEEE) in

1987 as the IEEE standard 1076, initially was mandated by the U.S. Department of Defense.

Dataquest Perspective

Dataquest believes that this consortium was formed in response to the creation of OVI a year ago for the purpose of establishing

Cadence's Verilog HDL as the preeminent hardware description language for the EDA industry.

This is, in essence, an attempt to ensure that

VHDL will indeed overshadow Verilog HDL.

Given the fact that, to date, very few vendors have rushed to embrace OVI and Verilog HDL, this consortium may well prove to be a successful move by Cadence's competitors. In fact,

Dataquest believes that VHDL most likely will move ahead by the third quarter of this year, when Mentor Graphics ships its own VHDLbased simulation and design system.

Dataquest estimates the installed base of VHDL users versus Verilog HDL users to be roughly equivalent, at about 2,000 users each. In

Europe, VHDL has a high profile, as opposed to the EDA market in Japan, where Verilog has a higher profile than Verilog HDL. In North

America, the market is split, although Verilog has greater momentum within its user base because Verilog users are more seasoned in terms of Verilog HDL-based top-down design.

But at the same time, the momentum among potential hardware description languages is shifting toward VHDL because most of the people in that market do not see enough third-party support, especially for Verilog HDL-based logic simulators.

Finally, Dataquest firmly believes that unless

OVI becomes aggressive and facilitates a Verilog done simulator industry, Verilog's future remains uncertain. It is not likely that the major EDA or

ASIC vendors that have lined up behind VHDL will offer a Verilog simulator because doing so would, in essence, provide an endorsement for the Verilog HDL. It is much more likely that smaller ASIC and EDA vendors will be more inclined to offer a Verilog HDL-based simulator.

But again, this scenario is not likely to happen until OVI creates an infrastructure for a clonesimulator industry to quickly develop. •

By Ron Collett

In Future Issues

Topics of interest to the CAD/CAM/CAE industry will appear in upcoming issues of Dataquest

Perspective, including the following:

• A look at scalar compute power across computing platforms ranging from PCs to supercomputers

• An in-depth analysis of CADDS 5, Computervision's latest update of its CAD/CAM software

• More "News and Views"

For More Information . . .

Managing Editor Patrick Waurzyniak (408) 437-8132

On the topics in this issue CAD/CAM/CAE (408) 437-8132

On related subjects Client Inquiry Center (408) 437-8671

About other Dataquest publications Conferences (408) 437-8245

About your subscription Customer Service (408) 437-8402

Via fax request Fax (408) 437-0292

The content of this report represents our Interpretation and analysis of information generally available to die public or released by responsible Individuals In the subject companies, but is not guaranteed as to accuracy or comfdeteness. It does not contain material provided to us In confid«ice by our clients. Individual companies reported on and analyzed by

Dataquest may be clients of this and/or other Daiaquest services. This informaaon is not furnished In connecticm with a sale or offer to sell securities or In connection with the soUcitstion of an offer to buy securities. This firm and Its parent and/or their officers, stockholders, or members of theJr fiunllles may, from time to time, have a long or short position

In the securfdes mentioned and may sell or buy such securities.

©1991 Dataquest Incorporated / 1290 Ridder Park Drive, San Jose, CA 95131-2398 / (408) 437-8000 / Fax (408) 437-0292

0010745

$13 i-t

?r

r-t-

I. ff.

r>

Forecasts

Electronic Design Automation Applications

Library Copy

DO NOT REMOVE!

CAD/CAM/CAE

INFORMATION RESOURCE CENTER

DATAQUEST INCORPORATED

1290 Ridder Park Drive

San Jose, CA 95131 2398

Forecasts '''"''''""'

Electronic Design Automation Applications

Dataoyest CAD/CAM/CAE

Source:

Dataquest

i i

Published by Dataquest Incorporated

The content of this report represents our interpretation and analysis of information generally available to the public or released by knowledgeable individuals in the subject industry, but is not guaranteed as to accuracy or completeness. It does not contain material provided to us in confidence by our clients.

Printed in the United States of America. All rights reserved. No part of this publication may be reproduced, stored in retrieval systems, or transmitted, in any form or by any means—mechanical, electronic, photocopying, duplicating, microfilming, videotape, or otherwise—without the prior permission of the publisher.

© 1991 Dataquest Incorporated

September 1991

i

Table of Contents

Forecasting Methodology.

Page

1

List of Figures

Figure Ps^e

Figure 1 Information Compendium 2

Index of Tables

Topic

ELECTRONIC DESIGN AUTOMATION

Topic Page

Table 26 NA Technical Workstation 28

ELECTRONIC CAE

Table 46 NA Technical Workstation 48

i

Forecasts—Electronic Design Automation Applications

Topic Page Topic Page

©1991 Dataquest Incorporated September—Reproduction Prohibited

ForecastsElectronic Design

Automation Applications

Forecasting Methodology

Fundamental to the way Dataquest conducts its research is an underlying philosophy that says the best data and analyses come from a wellbalanced program. This program includes balance between primary and secondary collection techniques; balance between supply-side and demand-side analysis; balance between focused, industry-specific research and coordinated, "big-picture" analysis aided by integration of data from the more than 25 separate high-technology industries Dataquest covers; and balance between the perspectives of experienced industry professionals and rigorous, disciplined techniques of seasoned market researchers.

Dataquest also analyzes trends in the macroenvironment, which can have major influences on both supply-side and demand-side forecasting. In addition to demo^^phics, analysts look at GNP growth, interest rate fluctuation, currency fluctuation, business expectations, and capital spending plans. In the geopolitical arena, the group looks at trade issues, political stability or lack thereof, tariffs, nontariff barriers, and such factors as the effert on Europe of the events of 1992.

Figure 1 is a pictorial summary of the information compendium offered by the Dataquest

CAD/CAM/CAE group.

Dataquest's revenue and shipment estimates are based on the following sources:

• Information supplied by company management or gathered from publicly available published sources

• End-user surveys

• Information supplied by other Dataquest industry services relating to components/subsystems of CAD/CAM/CAE

• Information provided by OEMs or resellers of the manufacturers' products

• Senior staff estimates based on reliable historical data and in-depth understanding of a company's products and strategies

The CAD/CAM/CAE industry market estimates and forecasts are derived using the following research techniques:

• "Bottom-up" aggregation—This method involves adding all relevant vendor contributions to arrive at total market estimates for all historical data.

• Segment forecasting—^For each application segment tracked by the CAD/CAM/CAE group, individual forecasts are derived following the basic information model defined previously. Sf>ecificaUy, each design phase covered within each application is segmented by channel, product, region, and platform. In this way, each application segment incorporates its own set of unique assumptions.

• Demand-based analysis—^Market growth is tracked and forecast in terms of the present and anticipated demand of current and ftjture users. This requires the development of a total available market (TAM) model and a satisfied available market figure to accurately assess the levels of penetration. Installed base and rates of product retirement are also evaluated. In addition, Dataquest analysts factor in the acceptance or ability for users to consume new technology.

• Capacity-based analysis—This method involves identifying future shipment volume constraints. These constraints, or "ceilings," can be the result of component availability, manufacturing capacity, or distribution capacity. In any case, constraint is capable of keeping shipments below the demand level.

rigare 1

Information Compendium

Forecasts—Electronic Design Automation Applications

Support

Mechanisms

Information

Tools

i

Source: Dataquest (September 1991)

©1991 Dataquest Incorporated September—Reproduction Prohibited

TITLE:

APPLICATION:

REGION:

PLATFORM:

1

Final History and Forecast

Electronic Design Automation

Worldwide

All Platforms

e

•S

§

1

a

a

Sf

•o

a

1

I

•o

I cr

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 43,729 55,216 73,417 84,443 95,712 111,530 132,580 151,210 17

Workstation Shipments 46,472 57,723 74,451 87,634 98,662 114,290 134,940 153,080 17

CPU Installed Base 78,045 130,707 196,813 265,443 339,497 402,920 494,010 605,320 74

Workstation Installed Base 86,864 142,384 209,988 282,305 359,534 427,300 521,730 635,800 77

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

(Thousands of U.S. Dollars)

78.4 60.3 62.8

Hardware-Only ASP

17.0 14.1 11.3

52.4

12.0

50.9

11.6

50.6

11.1

49.5

10.7

49.0

10.6

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 1,149 1,109

1,282

CPU Revenue 834 745 960

Workstation Revenue 136 138 62

Peripheral Revenue (Turnkey) 179 226 260

Software Revenue 544 826

912

Bundled 307 305 344

U n U n d l e d 237 521 568

Service Revenue 186 288

364

Total Revenue 1,879 2,224 2,558

Increase over Prior Year 35X 18% 15%

1,417

1,159

73

185

1,064

456

608

424

2,905

14%

1,488

1,244

69

175

1,156

396

759

524

3,168

9%

1.567

1,115

111

341

1,255

420

836

564

3,385

7%

1,748

1,249

105

394

1,458

464

994

654

3,860

14%

1,923

1,377

92

453

1,724

504

1,220

769

4,416

14%

TITLE:

APPLICATION:

REGION:

PLATFORM:

Final History and Forecast

Electronic Design Automation

Worldwide

Technical Workstation

®

S"

fS

•g

2

s

0

3

S a

if

t3

H

3

sr

<B

TD

^ c n

0

3

•V

3

E g-

1

19S6 ssss

UNIT SHIPMENT DATA (Workstation Shipments)

1987

====

1988 sssr

1989

====

1990

====

1991

====

1992

====

1993

===s

23,635

63,035

37,024

37,024

95,431

95,431

40,474

40,474

134,549

134,549

51,780

51,780

167,860

167,860

70,780

70,780

226,810

226,810

93,960

93,960

309,610

309,610

1

1

4

4

56.9

14.5

55.9

15.6

53.7

14.1

50.9

12.9

49.1

12.3

REVENUE DATA (HiUions of U.S. Dollars)

677

498

0

814

688

0

Peripheral Revenue (Turnkey) 91

134 179 126

817

Bundled 212

411

406

307

1,938

Increase over Prior Year

34X

41X 29X 23X

884

755

0

129

882

348

533

411

2,177

12X

984

712

0

272

958

351

607

451

2,393

10%

1,192

864

0

328

1,187

408

778

550

2,928

22%

1,437

1,043

0

394

1,507

462

1.044

681

3,625

24X

TITLE:

APPLICATION:

REGION:

PLATFORM:

Final History and Forecast

Electronic Design Automation

Worldwide

Host/Server

I

O

I

I

1986 1987 1988 1989 1990 1991

1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 1,532 1,694 2,866 3,815 4.133

4,370 4.320 4,010

Workstation Shipments 4,276 4,200 3,900 7,005 7,083

7,130 6,670 5,880

CPU Installed 8ase 4,587 5,928 8,331 11,650 14,612 18,350 21,670 24,790

Workstation Installed Base 13.406 17,606 21,506 28,512 34,649 42,720 49,400 55,280

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 442.8 277.0 327.6 300.9 124.6 120.3 117.5 116.1

Hardware-Only ASP 248.1 275.4 87.5 74.5 72.5 67,9 63.1 58.6

REVENUE DATA (HilUorp Of U.S. Dollars)

Hardware Revenue 427 426 340 339 312 287 263 227

CPU Revenue 220 207 217 224 218 135 121 103

Workstation Revvnyie 136 138 62 73 69 111 105 92

Peripheral Revenit (Turnkey) 72 82 61 42 25 41 38 32

Software Revenue 118 114 103 79 78 97 90 73

BiMvdled 54 41 41 27 27 45 37 29

Ur^xndled 63 72 62 52 51 52 52 44

Service Revenue 79 89 95 88 82 82 75 64

Total Revenue 624 629 539 507 472 465 428 364

Increase over PrlAr Year 11X IX -14X -6X -7X -IX -8X -15X

TITLE:

APPLICATION:

REGION:

PLATFORM:

Final History and Forecast

Electronic Design Autoniation

Worldwide

Personal Computer

©

I

o

»

I i

s

SP

•o

H

3

c

I

§•

E>

•O

3

W.

cr

I

1986 1987

1988

1989

1990 1991 1992

1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 29,819 37,729

46,916

Workstation Shipments 29,819 37,729

46,916

43,604

43,604

158,362

158,362

51,105

51,105

190,336

190,336

55,380

55,380

216,720

216,720

57,490

57,490

245,530

245,530

53,240

53,240

270,920

270,920

4

4

28

28

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S.

Dollars)

28.4 11.1

14.8 17.5 18.8

6.6 4.7 5.3 5.2

5.0

19.3

5.0

19.1

4.9

18.4

4.8

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 235 187 264 264 291 296 293 258

CPU Revenue 219 177 244 246 271 269 264 231

Workstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 16 10 20 18 21 27 29 27

Software Revenue 133 166 155 168 196 200 182 144

Biodled 41 18 11 18 21 23 18 12

Untwidled 92 148 144 150 175 177 164 132

Service Revenue 19 22 24 29 31 31 29 24

Total Revenue 387 375 444 461 519 527 503 427

Increase over Priop Year 120X -3X 19X 4X 12X 2X -5X -15X

TITLE:

APPLICATION:

REGION:

PLATFORM:

Final History and Forecast

Electronic Design Automation

North America

All Platforms

®

o

»

R

0 a

"U

H

I n

"O

c

I

r>

c.

0

3

TJ

3

1

1

1986

====

UNIT SttlPHEHT DATA (Workstation

Shipments)

CPU Shipments

Uorltststion Shipments

CPU Installed Base

Uorkstation Installed Base

24,338

25.396

48,832

53,865

1987

====

1988

====

1989

=s==

1990

====

1991

====

1992

====

1993

====

24,379

25,412

71,302

77,569

37,731

37,891

103,906

110,629

37,140

38.259

131,021

139,227

40,352

41,806

160,158

170,588

46,500

47,800

179,880

191,870

54,260

55,250

212,290

225,790

61,290

62,010

251,990

266,690

7

7

30

31

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 67.0 58.1 66.9

Hardware-Only ASP 15.7 13.2 10.2

58.5

12.0

59.3

12.0

56.4

11.5

53.2

11.3

51.3

11.3

HardMare Revenue

534 420

CPU Revenue

Workstation Revenue

395

55

273

54

Peripheral Revenue (Turnkey) 84

94

Software Revenue

Bundled

Unbmdled

Service Revenue

Total Revenue

Increase over Pri^pr Year

261

105

156

103

898

16%

346

81

266

128

895

-OX

534

411

20

104

388

115

273

159

1,081

21%

557

466

29

62

447

132

315

195

1,198

11X

586

491

35

60

498

122

376

245

1,328

11X

597

450

37

110

522

135

386

250

1,368

3X

661

501

32

127

574

150

424

278

1,512

11X

719

548

25

145

667

160

507

318

1,704

13X

TITLE:

APPLICATION:

REGION:

PLATFORM:

Final History and Forecast

Electronic Design Automation

North America

Technical Workstation

®

*-*

^

^

1

P n

0

3 a

•o

5r

7

13 i

1

1 s

1 ff o*

1

1986

===s

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments

Workstation Shipments

6,561

6,561

CPU Installed Base

Workstation Installed Base

16,443

16,443

1987 srss

1988

====

1989

====

1990

====

1991

====

1992

£===

1993

====

7,172

7.172

22,992

22,992

10,542

10,542

31,943

31,943

15,602

15,602

44,517

44,517

17,312

17.312

61.721

61,721

22,430

22,430

73,410

73,410

31,760

31,760

98,960

98,960

42,970

42,970

135,720

135,720

1

18

CALCULATED AVERAGE SYSTEM PRICE

. Dollars)

Turnkey ASP

60.6

51.2 58.8

Hardware-Only ASP

23.3

22.0 19.2

55.0

15.0

56.6

15.5

54.1

14.1

51.9

13.0

50.6

12.4

REVENUE DATA (Millions of U.S. Dollars)

HardHsre Revenue

212

182

CPU Revenue

Service Revenue

173

132

Workstation Revenue

Softuare Revenue

0

0

Peripheral Revenue (Turnkey) 39

50

141 215

Bundled

83

68

Unbundled

58

147

51 77

Total Revenue

403

475

Increase over Prior Year

11X 18X

269

198

0

70

272

107

165

106

646

36X

307

268

0

39

333

128

205

143

782

21X

338

294

0

44

380

117

264

194

912

17%

363

277

0

86

410

122

288

203

976

7X

453

348

0

105

486

140

345

239

1,178

21X

552

425

0

126

604

153

450

287

1,443

23X

TITLE:

APPLICATION:

REGION:

PLATFORM:

Final History and Forecast

Electronic Design Automation

North America

Host/Server

6

*-*

S"

s

•i

a

R

o

a

I

5f

•o

a

•8

I

c r,

I

g-

3

I

1986

UNIT SHIPMENT DATA (Uorkstation Shipments)

1987 1988 1989 1990 1991 1992 1993

797 1,400 1,948

3,066

5.857

14,063

1,867 1,900 1,810

3,320 3,200 2,810

7.322 8.590

9,880

17,753 20,580 23,390

1,620

2,330

11,020

25,720

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 351.6 179.5 211.9 193.4 250.8

Hardware-Only ASP 219.8 245.4

85.4 69.3 70.0

222.3

65.3

198.5

59.6

REVENUE DATA (Mi I lions of U.S. Dollars)

Hardware Revenue 201 166 142

CPU Revenue 105 71

90

Uorkstation Revenue 55 54 20

Peripheral Revenue (Turnkey) 41 42 32

Software Revenue 55 47

37

Bundled 13 8

6

Unbundled 43 39 31

Service Revenue 43 42

41

Total Revenue 300 255

221

Increase over P H p r Year 4X -15X

-13%

149

100

29

20

37

4

34

42

228

3X

141

94

35

11

31

5

26

39

211

•ex

123

65

37

20

35

13

22

36

193

-8%

105

55

32

18

26

9

17

30

161

-16X

173.7

54.6

85

45

25

15

18

7

11

24

127

-21%

TITLE:

APPLICATION:

REGION:

PLATFORM:

8

Final History and Forecast

Electronic Design Automation

North America

Personal Computer

n

I

3

H

a

Sf

•a

3

7) n

•a

I

§

9

"O

3

Er a

I

1986 1987 1988 1989

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP 20.0 20.4 98.7

Hardware-Only ASP

7.0 4.2 4.6

28.3

5.1

1990

24.7

5.0

1991

UNIT SHIPMENT DATA (Uorkstation Shipments)

CPU Shipments 16,948 16,411

25,789 19,591 21,173

WorkstBtion Shipments 16,948 16,411 25,789 19,591 21,173

CPU Installed Base 29,815 45,141 67,689 80,646 91,114

Uorlcstation Installed Base 29,815 45,141 67,689 80,646

91,114

22,170

22,170

97,880

97,880

20,680

20,680

103,440

103,440

16,700

16,700

105,240

105,240

1

1

23.9

5.0

1992

24.6

4.9

1993

23.7

4.9

REVENUE DATA (Hi 11 ions of U.S. Dollars)

Hardware Revenoe 120 72 123 101 107 111 102 82

CPU Revenoe 117 71 122 98 103 108 98 78

Uorkstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 3 2 1 3 5 3 4 4

Software Revenue 65 84 80 77 86 77 62 46

BiXKlled 10 4 3 1 0 0 0 0

Unbmdled 55 79 77 76 86 77 62 46

Service ReverHje 9 9 12 10 12 11 10 7

Total Revenue 195 165 215 188 205 200 174 135

Increase over Prf«r Year 62X -15% 30% -12X 9X -3X -13% -22%

TITLE:

APPLICATION:

REGION:

PLATFORM:

Final History arxl Forecast

Electronic Design Automation

Europe

All Platforms

@

K

1

s

a ce

1

s

I

Jp

I

•o

3

O

1

1 cr

1

1986

====

UNIT SHIPMENT DATA (Workstation

Shipments)

CPU Shipments

11,423

Workstation Shipments

CPU Installed Base

Workstation Installed Base

11,838

16,721

17.848

1987

====

1988

====

19,178

19,682

35,606

37,240

21.844

22,032

56,228

58,066

1989

====

1990

====

1991

====

1992

====

1993

====

27,696

28,844

80,435

83,478

32,835

33,654

107,472

111,571

38,770

39,550

132,690

137,600

44,750

45,460

165,480

171,250

48,250

48,810

201,920

208,350

5

5

24

24

CALCULATED AVERAGE SYSTEM PRICE

. Dollars)

Turnkey ASP

70.7 38.0 44.1

Hardware-Only ASP

14.9 12.8 13.6

48.6

12.6

48.6

11.5

50.6

10.9

50.6

10.4

49.8

10.5

REVENUE DATA (Hillions of U.S. Dollars)

Hardware Revenue

CPU ftevenoe

283

226

312

221

Workstation Revenue

15 30

Peripheral Revenue (Turnkey) 41

61

Software Revenoe

Bundled

Unbundled

Service Revenue

Total Revenue

Increase over Pritfr Year

153

98

54

39

474

80%

281

83

198

94

687

45X

400

329

14

58

261

82

179

116

777

13X

450

381

28

41

277

113

164

141

869

12%

474

413

22

39

307

104

204

177

958

10%

562

399

41

122

334

103

231

205

1,102

15%

629

444

40

145

406

119

287

245

1,280

16%

700

488

35

176

498

142

356

298

1,496

17%

TITLE:

APPLICATION:

REGION:

PLATFORH:

10

Final History and Forecast

Electronic Design Automation

Europe

Technical Workstation

©

I — * o

M

I

3

8 s

H

a

if

•o

H

B

I

n

T3

I

c

I

3

13

I a

1986

1987 1988 1989

REVENUE DATA (Hiltions of U.S. Dollars)

Hardware Revenue 153 140 212

244

CPU Revenue 126 102 169

220

Workstation Revenue 0 0 0

0

Peripheral Revenue (Turnkey) 27 38 43

24

Software Revenue 92 193 191

211

Unb(*»dled 14 119 119

105

Total Revenue 265 391 479

559

Increase over Prior Year 90% 47X 23X

17%

1990

UNIT SHtPMEHT DATA (Workstation Shipments)

10,669 17,453

27,276

27.276

11,777

11,777

38,686

38,686

15,520

15,520

49,500

49,500

20,030

20,030

66,400

66,400

25,790

25,790

89,110

89,110

1

1

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP

81.9 56.0 53.2

57.5

Hardware-Only ASP

26.7 23.5 22.3 15.5

54.3

17.0

52.0

15.4

50.0

14.1

48.9

13.4

258

236

0

22

230

94

135

139

627

12%

1991

346

247

0

99

245

87

158

167

757

21%

1992

416

293

0

123

320

105

215

208

944

25%

1993

513

358

0

156

428

132

296

265

1,206

28%

TITLE:

APPLICATION:

REGION:

PLATFORM:

11

Final History and Forecast

Electronic Design Automation

Europe

Host/Server

©

>-*

D

S

R

8

1

a

1

I

a

9

1

O"

1

1986

1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments

Workstation Shipments

158

573

540

CPU Installed Base

Workstation installed Base 1.667

345

849

883

2.517

922

1.110

1,789

3,627

1,140

2,288

2,872

5,915

1,136

1,955

3,798

7,897

1,230

2.010

4,970

9,880

1,260

1,960

6,080

11,840

1,200

1,760

7,170

13,610

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP

Hardware-Only ASP

516.7

389.0

392.4

284.5

252.6

93.7

209.5

93.9

293.6

81.0

295.7

76.1

285.2

72.1

266.3

67.1

REVENUE DATA (Millions of U.S.

)ollars)

Hardware Revenue 62

39

CPU Revenue

Workstation Revenue 15

Software Revenue

Bundled

Untx#KJled

Service Revenue

Total Revenue

Increase over Prior Year

21

6

15

14

98

2X

102

54

30

32

5

27

26

161

64X

102

76

14

12

21

6

15

31

153

-4%

115

76

28

11

12

4

8

28

154

1%

100

73

22

6

16

6

10

28

144

-6X

96

42

41

12

22

13

9

28

146

1%

91

40

40

11

20

12

8

27

138

-5%

79

34

35

10

17

10

7

23

120

-13%

TITLE:

APPLICATION:

REGION:

PLATFORM:

12

Final History and Forecast

Electronic Design Automation

Europe

Personal Computer

@

§

1 i a

SP

•o

H

3 ft

•o

I

E-

3

"O

3

E

IT

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 7,825 13,954

13,761

Workstation Shipments 7,825 13,954

13,761

CPU Installed Base 10,288 24,054

36,986

Workstation Installed Base 10.288 24,054

36,986

15,796

15,796

50,286

50,286

19,922

19,922

64,988

64,988

22,030 23,470 21.260

22,030 23,470

21,260

1

1

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP

Hardware-Only ASP

7.5 4.7

6.0

12.9

5.3

17.1

5.2

19.5

5.2

18.4

5.1

17.0

5.0

REVENUE DATA (Hit I ions of U.S. Dollars)

Hsrduare Revenue 67 70 86 92 115 120 122 107

CPU Revenue 61 65 84 85 105 110 111 97

Workstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 6 5 3 7 11 11 11 10

S o f t w r e Revenue 39 57 49 55 61 68 65 53

Buylled 14 5 5 4 3 4 2 0

Unbundled 25 52 44 51 58 64 63 53

Service Revenue 5 10 9 9 10 10 10 9

Total Revenue 111 136 144 156 186 198 197 170

Increase over Prior. Year 299% 22X 6X 8X 20X 6X -OX -14X

TITLE:

APPLICATION:

REGION:

PLATFORM:

13

Final History arxi Forecast

Electronic Design Automation

Asia

All Platforms

©

t^

o

K

•g

8 g o

O

1

3

a

ff

•o

ft

1

k

t3

1

3

?

1 cr

1

1986

====

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments

Workstation Shipments

CPU Installed Base

Workstation Installed Base

7,098

8,209

10,905

12,841

1987

====

10,629

11.558

21,352

24,219

1988

====

13,029

13,686

33,643

37,181

1989

====

1990

====

18,026

18,838

49,603

54,014

20,649

21,186

65,942

70,067

1991

====

1992

====

1993

====

23,390

23,870

82,390

88,270

28,700

29,110

104,120

110,700

34,160

34,480

132,490

139,670

1

1

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S . Dollars)

109.4 98.5 84.2

Turnkey ASP

Hardware-Only ASP 21.4 18.3 11.1

51.4

10.6

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue

CPU Revenue

Workstation Revenue

293

184

58

355

236

51

Peripheral Revenue (Turnkey) 51

68

Software Revenue

Bundled

126

101

194

140

Unbundled

25

54

Service Revenue

63

Total Revenue

Increase over Prior Year

39

458

36X

612

34X

332

208

26

97

257

145

112

85

674

10%

388

294

14

81

334

209

125

83

806

19X

47.9

10.5

404

321

9

74

343

169

174

99

846

5X

47.4

10.4

377

246

26

105

387

178

209

102

866

2X

46.6

10.0

46.6

9.7

414

272

25

117

460

191

269

122

996

15X

443

295

23

126

532

196

336

141

1,117

12X

TITLE:

APPLICATION:

REGION:

PLATFORM:

14

Final History and Forecast

Electronic Design Automation

Asia

Technical Workstation

«

•o f?

3

?

T

Jb

I

I

ff o

R

I

•O

3

E a

3

I

1986 1987

1988

1989

1990 1991

1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 1,948 3,533

5,666 10,191 10,909

Workstation Shipments 1,948 3,533

5,666 10,191 10,909

12.584 22,183 32,241

32,241

13,200

13,200

42.790

42.790

18.080

18.080

58,620

58.620

23.850

23.850

80.850

80,850

1

1

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 128.0 110.8 78.9 58.0

56.6

Hardware-Only ASP 26.0 24.0 13.4 11.3

13.7

55.1

12.5

51.1

11.5

48.7

10.9

REVENUE DATA (Millions of U.S. Dollars)

Harduare Revenue 109 168 191 255

281

CPU Rever*ie 85 123 127 193

219

Workstation Revenue 0 0 0 0

0

Peripheral Revenue (Turnkey) 24 45 65 62

62

Software Revenue 59 137 189 271

269

Bundled 49 103 112 176

136

Unbmdled 9 34 77 94

133

Service Revenue 15 42 60 58

76

Total Revenue 183 346 440 584

625

Increase over Prior Year 32X 90% 27X 32%

7X

266

181

0

85

298

141

157

78

642

3%

312

214

0

98

370

161

209

99

781

22%

357

248

0

109

459

174

285

122

938

20X

TITLE:

APPLICATION:

REGION:

PLATFORM:

15

Final History and Forecast

Electronic Design Automation

Asia

Host/Server

O

>-*

S*

I

n

a

if

•o

n

3

B:

CT

R a

1986 1987 1988 1989 1990 1991 1992

1993

UNIT SHIPMENT DATA (Workstation Shipments)

1,080 980

Workstation Shipments 1,554 1,435 1,149 1,467

1,584 1,590 1,490 1,300

CPU Installed Base 1,058 1,561 2,040 2,634 3,147 4,340 5,130 5,840

Workstation Installed Base 2,994 4,428 5,578 7,045 7,271 10,220 11,710 13,020

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 529.3 344,7 518.2 496.9 86.7 82.6 81.0 78.9

Hardware-Only ASP 264.2 344.2 72.8 50.2 57.0 52.9 49.5 46.6

REVENUE DATA (Hi 11 tons of U.S. Dollars)

Hardware Revenue 139 146 89 67 62 57 52 45

CPU Revenue 61 75 46 43 46 23 21 18

Workstation Revenue 58 51 26 14 9 26 25 23

Peripheral Revenue (Turnkey) 20 20 16 11 8 7 6 5

Software Revenue 40 34 44 29 30 38 41 35

Bundled 35 28 30 19 16 18 14 10

Unbundled 5 6 14 10 15 20 27 25

Service Revenue 19 19 21 17 13 15 15 13

Total Revenue 198 199 154 113 105 110 108 93

Increase over Prior Year 15X OX -23X -26% -7X 4X -2X -14%

TITLE:

APPLICATION:

REGION:

PLATFORM:

16

Final History and Forecast

Electronic Design Automation

Asia

Personal Computer

1

o

»

3

o

S

fi

o.

ff

•V

ft

3

I

•o c

1 e.

9

T)

3

a

1

1986

r===

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipnents

Uorkstation StiipM«nts

4,707

4,707

CPU Installed Base

UorkstBtion installed Base

6,193

6,193

1987

====

1988

=s==

6,590

6,590

12,659

12,659

6,871

6,871

19,019

19,019

1989

====

1990

====

1991

====

1992

====

1993

====

7,180

7,180

24,786

24,786

8,693

8,693

30,555

30,555

9,080

9,080

35.260

35,260

9,540

9,540

40.370

40,370

9.330

9.330

45.800

45.800

CALCULATED AVERAGE SYSTEM PRICE

DATA (Thousands of U.S . Dollars)

Turnkey ASP

35.4

15,1 14.8

Harduare-Only ASP

2.9 5.8 7.1

19.5

4.8

REVENUE DATA (Millions of U.S. Dollars)

Harduare Revenue

CPU Revenue

41

38

Uorkstation Revenue

0

Peripheral Revenue (Turnkey) 7

4

Software Revenue

0

27

23

Bundled

UnburxJled

Service Revenue

Total Revenue

46

39

17

10

5

77

9

15

3

68

Increase over Prior Year 185X

-12X

52

35

0

16

25

4

21

4

80

18X

66

58

0

8

35

14

20

9

109

36X

19.3

4.5

62

57

0

5

44

17

27

10

116

6X

19.2

4.5

54

41

0

13

50

19

31

9

114

-2X

19.1

4.4

50

37

0

13

49

16

33

8

107

-6X

18.5

4.4

41

29

0

11

38

12

27

6

85

-20X

TITLE:

APPLICATION:

REGION:

PLATFORM:

17

Final History arxj Forecast

Electronic Design Automation

Rest of World

All Platforms

e

«

t - *

I

I

a

•o

H

•o

I

c

9

I

i

o-

I

1986 1987 1988

1989

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 870 1,031 813

Workstation Shipments 1,029 1,072 841

CPU Installed Base 1,587 2,447 3,036

Workstation Installed Base 2,310 3,356 4,113

1,582

1,693

4,384

5,586

1990 1991

1992 1993

1,876

2,016

5,925

7.307

2,870

3,070

7,970

9,560

4,870

5,110

12,130

14,000

7,510

7,780

18,920

21,100

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

81.8 49.6 37.2

43.6 19.4 17.4

64.1

12.2

63.7

11.3

62.3

10.5

61.5

8.9

59.6

7.9

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 40 22 16 22 23 31 44 61

CPU Revenue 29 16 13 18 18 21 32 46

Workstation Revenue 8 3 2 3 3 6 8 9

Peripheral Revenue (Turnkey) 3 3 1 1 1 3 5 7

Software Revenue 4 4 6 6 7 12 19 26

Bundled 2 1 1 1 1 3 4 6

Unbundled 2 3 5 5 6 9 14 20

Service Revenue 5 3 4 5 4 6 9 12

Total Revenue 49 29 26 32 35 49 72 99

Increase over Prior Year 235X •40X -12X 26X 7% 42X 45X 38%

TITLE:

APPLICATION:

REGION:

PLATFORM:

18

Final History and Forecast

Electronic Design Automation

Rest of Uorld

Technical Workstation

«

o

B

S a

3

S"

^

3 i-

3: a-

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 428 211 267 471 475 630 900 1.340

Workstation Shipntents 428 211 267 471 475 630 900 1.340

CPU Installed Base 605 812 1.055 1.454 1.900 2.160 2.830 3.930 workstation Installed Base 605 812 1.055 1.454 1,900 2.160 2.830 3.930

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 62.4 51.8 28.2 47.5 48.3 46.1 44.2 43.1

Hardware-Only ASP 27,4 25.3 18.7 13.2 13.4 12.2 11.3 10.8

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 13 6 5 8 7 8 11 15

CPU Revenue 12 5 5 7 6 7 9 12

Workstation Revenue

Peripheral Revenue (Turnkey) 1

0

1

0

1

0

1

0

1 1

0 0 0 0

2 3

Software Revenue 2 2 3 3 3 6 10 16

Bundled 2 1 1 1 1 1 2 3

Unbundled 0 1 2 2 2 4 8 14

Service Revenue 2 1 2 2 2 3 5 7

Total Revenue 17 9 10 13 12 17 26 39

Increase over PrtOP Year 179X -46% 8% 30% -5X 4 U 53% 49X

TITLE:

APPLICATION:

REGION:

PLATFORH:

19

Final History and Forecast

Electronic Design Automation

Rest of World

Host/Server

n

•O

I

a

o

3

I

3

I

•a rf

3

®

H-*

o

V

B

1986

====

UNIT SMtPHEHT DATA (Workstation

Shipments)

CPU Shipments

103

Workstation Shipments

CPU Installed Base

Workstation Installed Base

263

415

1,137

1987

s=s=

46

87

316

1.225

1988

====

51

80

228

1.304

1989

====

1990

====

1991

====

1992

====

1993

====

73

184

286

1.488

83

224

345

1.728

130

330

450

2,040

170

410

580

2,450

210

480

760

2,930

CALCULATED AVERAGE SYSTEM PRICE

DATA (Thousands of U.S.

Dollars)

Turnkey ASP

Hardware-Only ASP

377.3

230.7

330.9

258.2

590.0

127.8

665.2

107.5

375.8

103.1

353.7

97.7

329.1

91.2

278.1

83.2

REVENUE DATA CHilMons of U.S. Dollars)

Hardware Revenue

CPU Revenue

Workstation Revenue

24

15

8

12

7

3

Peripheral Revenue (Turnkey) 2

2

Software Revenue airdled

Unbundled

Service Revenue

Total Revenue

Increase over Prior Year

1

1

0

3

28

264X

1

0

0

2

14

•49%

7

5

2

0

1

0

1

2

11

-27X

9

6

3

0

1

0

0

2

11

6%

9

5

3

0

1

0

0

2

12

4X

12

4

6

1

2

2

0

3

17

42%

15

5

8

2

2

2

0

3

21

25%

3

0

4

24

16%

17

6

9

2

3

TITLE:

APPLICATION:

REGION:

PLATFORM:

20

Final History and Forecast

Electronic Design Automation

Rest of World

Personal Computer

e

9

•fl

3

B: tr

I

a if

T3 f»

3 i

i

s

I

•3

I

c o

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHtPHENT DATA (Workstation Shipments)

CPU Shipments 339 773 495 1,038 1,317 2,110 3,800 5,950

Workstation Shipments 339 773 495 1,038 1,317 2,110 3,800 5,950

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 15.4 9 . 3 21.3 25.9 16.1 15.2 14.5 12.7

Hardware-Only ASP 6.6 4.8 6.2 5.3 5.0 4.9 4.9 4.8

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 2 4 3 6 7 1 0 18 29

CPU Revenue 2 4 3 5 7 10 18 27

Workstation Revenue 0 0 0 0 0 0 0 0

1 Peripheral Revenue (Turnkey) 0 0 0 0 0 0

Software Revenue 1 2 2 2 4 5

Buidled 0 0 0 0 0 0

6

0

1

7

0

Unbundled 1 2 2 2 4 5 6 7

Service Revenue 0 0 0 0 0 1 1 1

Total Revenue 4 6 5 8 11 16 25 36 increase over Prior Year 395X 53X -9X 61X 29% 44X 59% 46%

TITLE:

APPLICATION:

REGION:

PLATFORM:

21

Final History and Forecast

Electronic CAE

Worldwide

All Platforms

D

K

•g

n

VI

E s

a

5P

"O

n

1

f*

C

•o

I

Q

5

3

1

1

1986

====

UNIT SHIPMENT DATA (Workstation Shipments)

1987

s==r

1988

====

1989

====

1990

====

1991

====

1992

====

1993

====

39,135

118,066

53,536

54,535

158,456

161,527

60,916

61,735

203,070

207,154

70,070

70,800

246,230

251,430

83,450

84,020

304,650

310,780

94,510

94,920

375,330

382,230

4

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

48.7

9.8

44.4

9.6

44.4

9.4

44.5

9.3

45.0

9.4

REVENUE DATA (Hillior» of U.S. Dollars)

Peripheral Revenue (Turnkey) 100

Increase over Prior Year

31X

136

12%

627

465

15

147

737

601

25

111

595

18%

357

225

1.557

23%

761

641

22

99

647

179

468

280

1,688

8X

799

591

36

172

726

189

537

304

1,829

8X

931

685

34

212

879

223

656

373

2,182

19X

1,065

777

29

259

1,076

261

815

464

2,605

19X

TITLE:

APPLICATION:

REGION:

PLATFORM:

22

Final History and Forecast

Electronic CAE

Worldwide

Technical Workstation

@

D

K

•§

s

Sf

•o

S

n

•O

I

I a

"V

I cr

1986 1987

1988

1989 1990 1991 1992 1993

UNIT SHIPHeNT DATA (Workstation Shipments)

20,468

20,468

CPU Installed Base 17,859 26,741

39.192 55,755

72.765

Uorkstation Installed Base 17.859 26,741

39,192 55,755

72,765

26,130

26,130

89.760

89.760

37,280

37.280

119.710

119.710

50,160

50.160

163,010

163,010

2

2

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

62.1 52.2 50.1 52.5

25.7 23.2 18.8 14.4

50.1

16.2

48.5

14.6

46.3

13.4

45.3

12.7

REVENUE DATA (Hillions of U.S. Dollars)

Hardware Revenue 249 244 352 415 433 478 618 782

CPU Revenue 199 167 253 345 359 341 441 557

Uorkstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 50 77 99 70 74 137 177 225

Software Revenue 130 266 325 440 458 526 699 941

Bmdled 92 127 146 218 155 154 196 241

Ur*uidled 39 139 179 221 302 371 503 700

Service Revenue 43 94 128 164 221 246 320 420

Total Revenue 421 605 805 1,019 1.112 1.250 1.637 2.144

Increase over Prior Year 7X 43X 33X 27X 9X 12X 31X 31X

TITLE:

APPLICATION:

REGION:

PLATFORM:

23

Final History and Forecast

Electronic CAE

Worldwide

Host/Server

9

8 s

a

5f

•a

H n

•O

I

I

?

B: cr

S

1986 1987 1988 1989 1990 1991

1992

UNIT SHIPMENT DATA (Uorkstation Shipments)

CPU Shipments 671 920 1.474 1,719

1,887 2,000 1,970

Workstation Shipments 1,704 1,501 1,479 2,718 2.705 2,730 2,540

CPU installed Base 1,402 2,296 3,717 5,330 10,230

10,801 13,830

16.370

1993

1.830

2,240

11,710

18,610

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 384.3 199.9 193.5 174.9 87.4

Hardware-Only ASP 267.8 236.8 80.0 68.3 62.6

86.5

57.2

86.4

52.4

91.7

48.3

REVENUE DATA (Hillions of U.S. Dollars)

Hardware Revenue 188 198 142 135 119 102 94 82

CPU Revenue 100 97 88 81 84 51 46 40

Workstation Revenue 49 47 15 25 22 36 34 29

Peripheral Revenue (Turnkey) 38 54 39 28 12 15 14 13

Software Revenue 41 43 37 33 38 51 45 32

Bundled 7 7 5 4 7 20 16 12

UnbuTKiled 34 36 32 29 32 31 30 20 service Revenue 27 35 39 38 34 33 30 25

Total Revenue 256 276 218 206 191 187 169 139

Increase over

Prior

Year 34X 8X -21X -6X -7X -2%

-9% •^BX

TITLE:

APPLICATION:

REGION:

PLATFORM:

24

Final History and Forecast

Electronic CAE

Worldwide

Personal Computer

•o

n

3

if

§

I

3

K o.

I

ft

•o

I

i o

3

e

cr

1986 1987

1988 1989 1990

1991 1992

1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 21,035 19,063 23,530

Workstation Shipments 21,035 19,083

23,530

CPU Installed Base 35,519 53,394 73,191

Workstation Installed Base 35,519 53,394 73,191

31,969

31,969

97,370

97,370

38,561

38,561

123,588

123,588

41,940

41,940

147,840

147,840

44,210

44,210

174,710

174,710

42,520

42,520

200,610

200,610

2

2

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

29.5 12.2 19.1

6.8 4.7 5.2

REVENUE DATA (HilUons of U.S. Dollars)

Hardware Revenue 174 95

133

CPU Revenue 161 90

124

Workstation Revenue 0 0

0

Peripheral Revenue (Turnkey) 12 5

9

Software Revenue 88 81 94

Bundled 35 9

7

UnUJrttled 53 71

87

Service Revenue 14 12

15

Total Revenue 276 187

242

Increase over jP^rtor Year 93X -32%

29%

22.4

5.1

187

174

0

13

123

15

107

22

332

Z7X

20.4

4.9

210

198

0

12

151

17

134

24

384

16%

20.7

4.9

219

199

0

20

150

15

134

24

392

2%

20.2

4.8

219

198

0

21

135

11

123

22

376

-4%

19.9

4.8

201

180

0

21

103

8

95

18

322

-14%

TITLE:

APPLICATION:

REGION:

PLATFORM:

25

Final History and Forecast

Electronic CAE

North America

All Platforms

O

o

»

I a f

I c

§•

3

I

P-

I

1986 1987 1988 1989

1990 1991

1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 16,713 13,501 20,678 23,776 25.520

Uorkstatton Shipments 17,185 13,688 20,649 24,113 25,993

CPU Installed Base 36,097 48,107 64,761 81,061 96,841

Workstation installed Base 36,765 48,980 65,640 82,343

98,756

28,190

28,590

109,930

112,440

32,330

32,640

128,410

131,450

35,270

35,480

150,040

153,520

1

1

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

51.0 50.8 59.4 58.1 58.4

56.2

10.1

53.1

10.4

51.3

10.8

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 315 240 299 315 322 312 359 398

CPU Revenue 237 151 221 261 268 246 281 310

Workstation Revenue 24 22 7 10 12 10 9 8

Peripheral Revenue (Turnkey) 54 67 72 44 42 56 68 80

Software Revenue 139 178 229 291 325 345 387 465

Bulled 52 51 71 84 67 75 86 97

Unbi«lled 88 128 158 208 258 270 301 368

Service Revenue 52 71 91 113 135 136 159 192

Total Revenue 506 490 619 719 782 793 904 1,056

Increase over Prior Year 12X -3% 26% 16X 9X IX 14X 17%

TITLE:

APPLICATION:

REGION:

PLATFORM:

26

Final History and Forecast

Electronic CAE

North America

Technical Workstation

I

w

I

n

a if

•o on

•V

T3

I

I

3

1986

1987 1988 1989 1990 1991

1992 1993

UNIT SHIPMENT DATA (Uorkstation Shipments)

CPU Shipments 4,047 4,579 6,589

8,537

Uorkstation Shipments 4,047 4,579 6,589

8,537

CPU Installed Base 11,628 15,682 20,995

27,262

Uorkstation Installed Base 11,628 15,682 20,995

27,262

8,925

8,925

34,239

34,239

11,160

11,160

40,080

40,080

16,520

16,520

52,560

52,560

22,250

22,250

70,770

70,770

3

3

9

9

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP

50.7 43.1 52.2 55.4

56.4

Hardware-Only ASP

24.3 22.6 19.0 14.8

16.3

53.8

14.8

51.7

13.6

50.4

12.9

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 117 105 159 176 186 184 241 300

CPU Revenue 95 72 115 151 154 139 184 230

Workstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 22 32 44 25 32 45 57 70

Software Revenue 68 112 160 215 237 261 323 423

Bundled 41 45 67 83 65 66 79 92

Untumdled 27 67 93 132 172 196 245 330

Service Revenue 26 46 63 84 109 111 138 176

Total Revenue 211 263 382 475 531 556 702 898

Increase over Priof tear -10X 24X 46X 24X 12X 5X 26X 28%

TITLE:

APPLICATION:

REGION:

PLATFORM:

27

Final History and Forecast

Electronic CAE

North America

Host/Server

6

O

•S

I

ft

a

•a

•o

I

f) g-

o-

I

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPHEHT DATA (Workstation Shipments)

CPU Shipments 436 549 774 889 876 930 930 890

Workstation Shipments 908 736 744 1,227 1,349 1,330 1,240 1,100

CPU Instatled Base 880 1,412 2,151 2,974 3,583 4,420 5,120 5,780

Workstation Installed Base 1,549 2,286 3,030 4,256 5,498 6,930 8,170 9,270

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 295.4 134.2 185.1 152.8 167.2 159.7 153.2 146.8

Hardware-Only ASP 235.9 221.6 78.5 63.3 61.4 55.9 51.3 47.0

REVENUE DATA (Hi 11 ions of U.S. Dollars)

Hardware Revenue 107 98 76 , 65 58 48 45 39

CPU Revenue 54 41 42 39 39 29 27 24

Uorkstation Revenue 24 22 7 10 12 10 9 8

Peripheral Revenue (Turnkey) 29 34 27 17 7 9 8 7

Software Revenue 26 23 20 20 19 23 17 10

8i«lled 2 3 2 0 1 9 7 5

UnbuxJled 24 20 19 19 18 14 10 5

Service Revenue 18 20 20 20 17 16 14 11

Total Revenue 151 140 116 105 94 87 75 61

Increase over Prior Year 31X -7X -17X -9% -11% - A -13% -19%

T I T L E :

APPLICATION:

REGION:

PLATFORM:

28

Final History and Forecast

Electronic CAE

North America

Personal Computer

@

(-»

I ft

•o c

§•

9 ff

T3 f»

3 a i

I

a

ft

I

1986 1987 1988 1989 1990 1991

1992 1993

UNIT SHiraEHT DATA (Workstation Shipments)

CPU Shipments 12.230 8,373 13,316 14,350

Workstation Shipments 12,230 8.373 13,316 14,350

CPU Installed Base 23,588 31.013 41,615 50,825

Workstation Installed Base 23,588 31,013 41,615 50,825

15,719

15,719

59,019

59,019

16,100

16,100

65,430

65,430

14,880

14,880

70,730

70,730

12,120

12,120

73,490

73,490

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

20.4 32.9 88.0

32.1

Turnkey ASP

Hardware-Only ASP

7.3 4.3 4.5 5.1

59.7

5.0

59.1

5.0

58.5

4.9

57.9

4.9

REVENUE DATA (Millions of U.S. Dollars)

Hardyare Revenue 91 38 65

74

CPU Revenue 88 37 64

71

Workstation Revenue 0 0 0

0

Peripheral Revenue (Turnkey) 3 1 1

3

Software Revenue 46 44 49

57

Bundled 9 3 2

0

Unbundled 37 41 47

57

Service Revenue 7 6 8

8

Total Revenue 144 87 121

139

Increase over Prior Year 43% -39% 39%

15%

79

75

0

3

69

0

69

9

157

13X

80

78

0

2

61

0

61

9

150

•5X

73

70

0

3

47

0

46

7

127

-15%

59

56

0

3

33

0

33

5

97

-24%

TITLE:

APPLICATION:

REGION:

PLATFORM:

29

Final History and Forecast

Electronic CAE

Europe

All Platforms

I

n

i

73

I

I

I

a

I

}f

•V

3

I

1986 1987

1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 6,730 10,327

11,721

Workstation Shipments 6,835 10,476

11,694

CPU Installed Base 10,450 20.553

31,435

Workstation Installed Base 10,682 20,935 31,794

18,392

18,815

47,650

48,448

22,100

22,325

65,742

66,914

26,220

26,410

84,850

86,140

30,360

30,490

108,370

109,840

33,080

33,160

134,700

136,290

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

(Thousands of U.S. Dollars)

50.4 34.1 40.3

Hardware-Only ASP 15.4 13.0 13.6

49.7

10.1

REVENUE DATA (HiUions of U.S. Dollars)

Hardware Revenue 140 160 207

CPU Revenue 110 108 163

Uorkstatton Revenue 5 10 4

Peripheral Revenue (Turnkey) 26 42

40

Software Revenue 66 130

133

Bundled 42 40 43

Unbundled 24 90

90

Service Revenue 14 41 57

Total Revenue 220 332 397

Increase over Prior Year 56X 51X 20X

250

204

11

35

157

65

92

75

482

21%

47.1

9.4

264

227

7

30

180

55

125

94

537

12%

49.8

8.9

321

232

14

75

204

53

152

113

638

19%

50.0

8.6

367

263

12

92

266

63

203

143

777

22%

49.2

8.7

426

299

9

118

344

82

262

186

956

23%

TITLE:

APPLICATION:

REGION:

PLATFORM:

30

Final History

and

Forecast

Electronic CAE

Europe

Technical Workstation

®

l-t o

»

i

Vi

s

0

SP

"O

S

3

T

k c n

1

3

•0

cr

1

1986

====

UNIT SHIPMENT DATA (Workstation Shipments)

1987

====

1988

====

1989

====

1990

====

1991

====

1992

====

1993

=====

4.289

10,518

6,087

6,087

16,038

16,038

6,359

6,359

21,577

21,577

8.650

8,650

28.010

28,010

11.420

11,420

37.560

37,560

15.190

15,190

50.910

50,910

.

55.4

15.2

REVENUE DATA (Millions of U.S. Dollars)

Peripheral Revenue (Turnkey) 16

26

115

85

0

30

134

112

0

22

115

Unbundled 7

54

56

27X

306

23%

54.0

17.1

141

120

0

21

129

50

79

74

345

13%

51.8

15.5

198

136

0

63

148

47

101

95

441

28%

49.9

14.3

247

166

0

80

212

60

152

126

585

33%

48.8

13.5

318

212

0

106

300

79

221

171

789

35%

TITLE:

APPLICATION:

REGION:

PLATFORM:

31

Final History and Forecast

Electronic CAE

Europe

Host/Server

6

I

§

I

a

}f

T3

3

i

I

I

1986 1987 1988

1989 1990

1991 1992

1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipnients 49 181 489 528 524 510 470 410

Workstation Shipments 155 330 462 950 749 690 590 490

CPU Installed Base 160 340 825 1,337 1,825 2,290 2.700 3,070

Workstation Installed Base 392 722 1,184 2,135 2,997 3,580 4,170 4,660

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 683.0 506.9 144.7 113.4 172.8 179.9 172.3 165.1

HardMare-Only ASP 548.9 230.9 86.7 87.0 71.1 65.3 60.2 55.4

REVENUE DATA (Hi 11 ions of U.S. Dollars)

HardHare Revenue 27 48 45 47 38 32 27 22

CPU Revenue 19 24 34 29 29 14 12 9

Workstation Revenue 5 10 4 11 7 14 12 9

Peripheral Revenue (Turnkey) 3 13 7 7 3 4 4 3

SoftMare Revenue 9 14 10 6 9 11 9 8

Bundled 1 1 1 0 2 4 3 3

Unbundled 8 13 9 6 7 7 6 5

Service Revenue 2 9 13 12 12 11 9 7

Total Revenue 38 71 69 65 59 53 45 37

Increase over Prior Year -8X 85X -2X -6X -10X -9X -15X -19%

TITLE:

APPLICATION:

REGION:

PLATFORM:

32

Final History and Forecast

Electronic CAE

Europe

Personal Computer

6

O

I

• ^

i n

a

}f

•o

?

90

(I

•o

I

"fl

I a

1986 1987 1988

1989 1990 1991

1992 1993

UNIT SHIPMENT DATA (Uorkstation Shipments)

17,060

17,060

13,759 20,092

30,275 42,341 54,540

30,275 42,341 54,540

18,480

18,480

68,110

68,110

17,480

17,480

80,730

80,730

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

29.5 9.5 14.3 20.3

19.0

7.6 4.8 6.1 5.2 5.1

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 45 37 46 68

85

CPU Revenue 39 34 44 63

79

Workstation Revenue 0 0 0 0

0

Peripheral Revenue (Turnkey) 6 3 2 6

6

Software Revenue 22 24 28 35

42

Bundled 13 3 3 3

3

Unbundled 9 21 24 32

39

Service Revenue 4 5 5 7

7

Total Revenue 70 66 79 110

134

Increase over Prioir Year 212X -7X 20% 40X

21X

22.5

5.1

91

83

0

8

45

2

43

8

144

n

22.7

5.0

94

85

0

8

45

0

45

8

147

2%

.0

5.0

87

78

0

9

36

0

36

7

130

-11%

TITLE:

APPLICATION:

REGION:

PLATFORM:

33

Final History and Forecast

Electronic CAE

Asia

All Platforms

6 s-

B

•s f

I

I

I

I a

I

1986

s===

UNIT SHIPMENT DATA (Uorkstation Shipments)

CPU Shipments

Workstation Shipments

CPU Installed Base

5,065

5.396

7,337

Workstation Installed Base 7,614

1987

====

5,109

5,345

12,321

12,834

1988

====

1989

====

1990

====

1991

==r=

1992

====

1993

====

6.278

6.335

18,074

18,648

10,291

10,495

26,992

27.784

11.969

12,053

36,653

37,409

13,500

13,600

45.940

47.020

16,850

16,920

58,880

60,100

19,960

20,000

75,830

77,160

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S . Dollars)

Turnkey ASP

Hardware-Only ASP

66.8

16.2

68.4

17.3

51.0

10.7

41.4

8.6

34.2

8.8

32.3

9.1

33.5

8.9

34.9

8.8

Hardware Revenue

CPU Revenue

132

97

125

87

Workstation Revenue

Periptieral Revenue (Turnkey) 19

25

Software Revenue

Bundled

Unbundled

Service Revenue

Total Revenue

Increase over PrtW Year

16

53

40

13

16

201

56%

14

79

52

27

26

230

14%

113

75

3

35

91

44

47

32

236

2%

161

125

4

32

143

89

54

34

339

44X

163

135

2

26

137

56

80

49

349

3X

149

99

11

39

168

61

107

52

369

6%

178

118

11

49

212

72

140

66

456

24%

200

135

9

57

249

79

170

79

528

16%

•o

g-

3

"O

3

K cr

s

I

}p

a

3

Q

«

D u

S?

•9

TITLE:

APPLICATION:

REGION:

PLATFORM:

34

Final History and Forecast

Electronic CAE

Asia

Technical Workstation

1986 1987 1988

1989 1990

1991 1992 1993

UNIT SHIPMENT DATA (Uorkstation Shipments)

CPU Shipments 1,198 1,672 3,073

4,974

Uorkstation Shipments 1,198 1,672 3,073

4,974

CPU Installed Base 2,409 4,037 6,953

11,529

Uorkstation Installed Base 2,409 4,037 6.953

11,529

4,936 6,010 8,920

4,936 6,010

8,920

15,839 20,450 28,110

15,839 20,450 28,110

12,110

12,110

39,420

39,420

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 104.5 83.5 53.4 47.3

40.7

Hardware-Only ASP 26.5 23.8 13.3 11.5

14.5

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 55 60 75 101

102

CPU Revenue 44 41 51 78

82

Uorkstation Revenue 0 0 0 0

0

Peripheral Revenue (Turnkey) 11 19 24 23

21

Software Revenue 27 61 69 108

90

Bundled 23 45 40 74

39

Unbwidled 4 15 28 34

51

Service Revenue 7 20 25 23

37

Total Revenue 89 140 169 232

230

Increase o v w Prior Year 14% 57X 20X 38%

-IX

38.5

13.1

92

64

0

29

112

41

71

39

243

6X

37.0

12.1

126

87

0

39

156

57

99

54

335

38X

36.1

11.5

157

110

0

47

207

69

139

70

434

30X

TITLE:

APPLICATION:

REGION:

PLATFORM:

35

Final History and Forecast

Electronic CAE

Asia

Host/Server

6

9

i

I

I cr

I

3

o

I

n o.

if

•o ft

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 135 164 187 272 453 510 490 420

Uorkstetion Shipments 466 400 245 477 537 600 560 460

CPU Installed Base 262 425 609 867 1.138 1.720 2,130 2.480

Workstation Installed Base 538 939 1,183 1,660 1,894 2,790 3,360 3,820

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turr^ey ASP 693.1 319.5 425.1 357.5 57.3 55.5 53.3 51.4

Hardware-Only ASP 261.6 302.6 63.0 44.6 45.5 43.8 40.3 37.1

REVENUE DATA (HiUions of U.S. Dollars)

Hardware Revenue 40 47 18 19 20 18 17 14

CPU Revenue 20 28 10 11 15 6 5 4

Workstation Revenue 16 14 3 4 2 11 11 9

Peripheral Revenue (Turnkey) 4 5 4 4 3 2 1 1

Software Revenue 7 7 6 7 10 17 18 13

Bundled 4 4 2 3 3 6 5 3

Unbundled 2 3 4 4 7 10 13 10

Service Revenue 5 5 4 5 4 6 6 5

Total Revenue 52 58 28 31 35 40 41 32

Increase over Prior Year 65X 12X •53X 12X 11% 16% 2X -23X

TITLE:

APPLICATION:

REGION:

PLATFORN:

36

Final History

and

Forecast

Electronic CAE

Asia

Personal Computer

s a x>

c

I

•1

3

H

a

TJ

3

E

(T

^

•o i

•o

I

g-

1

1986

UNIT SHIPHEHT DATA <Uorksf»tion Shipments)

CPU Shipments

====

3.732

Uorkstation Shipments

CPU Installed Base

Workstation Installed Base

3.732

4.667

4.667

1987

====

1988

====

1989

=~=

1990

====

1991

====

1992

====

1993

====

3.273

3,273

7,859

7.859

3,018

3,018

10,512

10,512

5,044

5.044

14,595

14,595

6,581

6,581

19,676

19.676

6,990

6,990

23,780

23.780

7,440

7,440

28.630

28.630

7.430

7,430

33,920

33,920

CALCULATED AVERAt^ STSTEK PRICE

. Dollars)

Turnkey ASP

Hardusre'Only ASP

35.7

3.1

11.4

5.7

16.4

6.3

23.1

4.6

20.6

4.4

20.0

4.3

19.8

4.3

19.6

4.2

REVENUE DATA (HiUions of U.S. Dollars)

Hardware Revenue

37

CPU Revenue

33

Workstation Revenue

0

Peripheral Revenue {Turnkey)

Softuare Revenue

4

19

Bundled

Urdxndled

^Service Revenue

13

7

4

.Tbtal Revenue

60

Increase over Prior Year

208X

19

17

0

1

12

3

8

1

32

-47X

21

14

0

7

17

2

15

2

40

25X

41

36

0

5

29

13

16

7

76

91X

41

39

0

3

37

14

23

7

85

12X

39

30

0

9

40

13

27

7

86

n

36

27

0

9

38

11

28

6

80

•n

29

21

0

8

29

7

21

4

62

-22X

TITLE:

APPLICATION:

REGION:

PLATFORM:

37

Final History and Forecast

Electronic CAE

Rest of World

All Platforms

6

I-* i»t

D

^

(^

(A

§ o

1

a

JP

T}

I

1

I

D n go

1

(T

1

1986

====

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments

570

Workstation Stiiproents

CPU Installed Base

Workstation installed Base

695

897

1,021

1987

====

581

588

1,450

1,589

1988

==r=

1989

r===

453

456

1,830

1,984

1,078

1,112

2,752

2,952

CALCULATED AVERAGE SYSTEM PRICE

DATA (Thousands of U.S.

Dollars)

Turnkey ASP

51.8 74.2 24.9

Harduare-OnLy ASP 39.9

18.2 16.4

54.5

9.4

1990

====

1,327

1,363

3,833

4,075

49.3

8.5

1991

====

47.2

7.8

1992

====

45.8

6.8

1993

====

2,160

2,210

5,520

5,840

3,910

3,970

8,990

9,390

6,200

6,280

14,760

15,250

46.0

6.4

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue

CPU Revenue

Workstation Revenue

23

17

4

12

9

1

Peripheral Revenue (Turnkey) 2

2

Software Revenue

Bundled

1

0

Unbundled

Service Revenue

Total Revenue

1

3

26

2

1

1

2

16

Increase over Prior Year 284X •39X

8

7

1

1

4

1

3

2

13

-17%

11

10

1

1

4

0

3

2

18

31%

12

10

1

1

5

1

5

2

19

9X

17

14

2

2

9

1

7

3

29

50%

27

22

2

2

13

2

11

5

45

56%

40

33

3

4

18

3

16

7

65

44%

TITLE:

APPLICATION:

REGION:

PLATFORM:

38

Final History and Forecast

Electronic CAE

Rest of World

Technical Workstation g-

9

n

•o

I

§

TJ

3 fl

a

!f

•o it

3

S^

1986 1987 1988

1989 1990

1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 280 166 175 250 249 310 420 600

Workstation Shipments 280 166 175 250 249 310 420 600

CPU Installed Base 405 568 726 926 1,112 1,220 1,470 1.910

Workstation Installed Base 405 568 726 926 1.112 1,220 1,470 1.910

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 42.5 55.9 20.9 51.3 48.2 46.0 44.1 43.0

Hardware-Only ASP 28.0 25.0 18.6 12.6 12.9 11.7 10.8 10.2

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 8 4 3 4 4 4 5 7

CPU Revenue 8 3 3 4 3 3 4 6

Workstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 0 1 0 0 0 1 1 1

Software Revenue 1 1 2 2 2 4 8 11

Bi«Hed 0 1 1 0 1 1 1 1

UnbtFidled 0 0 1 1 1 3 7 10

Service Revenue 1 1 1 1 1 2 3 4

Total Revenue 10 6 6 7 7 10 15 22

Increase over Prior Year 168X -33X -lOX 21X -4X 48X 56% 45X

TITLE:

APPLICATION:

REGION:

PLATFORM:

39

Final History and Forecast

Electronic CAE

Rest of World

Host/Server

&

1

D

B

•i

§

i

a

I

5f

•o

g

-1

•o

I

1

1

9

1 cr

1

CPU Installed Base

Workstation Installed Base

1986

====

UNIT SHIPMENT DATA {Workstation Shipments)

CPU Shipments

Workstation Shipments

51

176

100

223

1987

====

27

34

119

258

1988

====

25

28

132

286

1989

====

30

65

151

351

1990

====

34

70

171

413

1991

====

60

110

210

530

1992

====

80

150

280

680

1993

====

110

190

370

860

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S.

Dollars)

Turnkey ASP

Hardware-Only ASP

437.5

256.1

270.2

199.7

463.9

118.7

449.0

98.0

136.4

90.4

130.3

82.3

124.9

75.6

119.7

69.3

REVENUE DATA CHiUions of U.S. Dollars)

13

Hardware Revenue

CPU Revenue

Workstation Revenue

8

4

6

Peripheral Revenue (Turnkey) 1

1

Software Revenue

0 0

Bundled

Unbundled

0

0

0

0

Service Revenue

Total Revenue

1

15

3

1

1

7

Increase over Prior Year 395X -54%

3

2

1

0

1

0

1

1

5

•28%

3

2

1

0

0

0

0

1

4

-19%

3

2

1

0

0

0

0

1

4

0%

4

2

2

1

1

1

0

1

6

50%

5

2

2

1

1

1

0

1

8

29%

1

1

0

2

7

3

3

1

10

27%

TITLE:

APPLICATION:

REGION:

PLATFORM:

40

Final History and Forecast

Electronic CAE

Rest of World

Personal Computer

I - * o

•§

3

o

I a

}f

"2

n

3

S"

ft

•a

I n

§•

3

TS

3

B: a

I

1986 1987 1988

1989

UNIT SHIPMENT DATA (Uork&tMlion Shipments)

CPU Shipments 239 388 253

797

Workstation Shipments 239 388 253

797

CPU Installed Base 392 763 972

1.675

Workstation Installed Base 392 763 972

1,675

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP

HardMare-Only ASP

11.5 36.8 29.4

6.5 4.8 6.1

32.2

5.2

REVENUE DATA (Millions of U.S.

Dollars)

Hardware Revenue

CPU Revenue

Workstation Revenue

Peripheral Revenue (Turnkey)

Software Revenue

Bundled

Urixjndled

Service Revenue

2

2

0

0

0

0

0

0

Total Revenue

Increase over P r l w Year

2

515X

2

2

0

0

1

0

1

0

3

42X

0

3

-9%

2

2

0

0

1

0

1 2

0

7

145%

0

2

0

4

4

0

1990

1,044

1,044

2,551

2,551

8.1

4.9

3

0

3

0

5

5

0

0

8

28X

1991 1992

0

5

1

22

68X

16

16

0

1

5

1993

1,790

1,790

4,090

4,090

3,410

3,410

7.240

7,240

5,490

5,490

12,480

12,480

8.0

4.9

4

0

4

9

8

0

0

0

13

52X

7.9

4.8

7.9

4.8

0

5

1

32

50X

26

25

0

1

5

TITLE:

APPLICATION:

REGION:

PLATFORH:

41

Final History and Forecast

IC Layout

Uorldwide

All Platforms e

•g

R o

1

a

I

if

•o fil

I n

•o

I

I

g-

I

1986 1987

1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

7,627

7.726

8,951 13,906 20.721

23,204

9,280

9,295

29,587

32.024

12,650

12,660

38,300

41,160

16,650

16,650

52,310

55,320

20,580

20,580

70,380

73,500

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 121.2 79.1 82.5 70.1 59.3

Hardware-Only ASP 35.4 26.5 21.8 17.9 17.5

55.6

15.1

52.8

13.4

52.2

12.4

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 140 141 144 163 191 204 231 258

CPU Revenue 100 97 110 147 175 144 166 188

Workstation Revenue 17 21 7 3 3 16 14 12

Peripheral Revenue (Turnkey) 23 23 28 12 13 44 50 57

Software Revenue 105 166 160 164 182 197 224 263

Bundled 47 36 36 35 38 53 57 62

Unbundled 58 130 124 129 145 144 166 201

Service Revenue 31 35 54 64 81 84 92 104

Total Revenue 276 342 359 390 454 484 547 625

Increase over l>rior Year 47X 24X 5X 9X 16X 7X 13X 14X

TITLE:

APPLICATION:

REGION:

PLATFORM:

42

Final History and Forecast

IC Layout

Worldwide

Technical Workstation

D

M

C

3

8

•i

s

I

3

if

•o

I

•fl

3

3:

tr

I

1986 1987 1988 1989 1990

146

137

0

10

165

36

128

67

378

25%

1991

164

124

0

40

174

48

126

70

409

8%

1992

195

149

0

47

198

53

145

79

473

16%

1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 1,838 2,389 3,367 6,212 8,008

Workstation Shipments 1,838 2,389 3,367 6,212 8,008

CPU InstsUed Base 3,188 5,526 8,707 14,431 22,553

Workstation Installed Base 3,188 5,526 8,707 14,431 22,553

11,310

11,310

31,090

31,090

15,280

15,280

44,700

44,700

19,390

19,390

62,520

62,520

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

(Thousands of U.S. Dollars)

97.8 93.0 97.0

66.3

Hardware-Only ASP

21.7 21.7 18.1 14.3

59.3

14.8

56.6

13.4

54.4

12.3

53.0

11.6

REVENUE DATA (Hi LI ions of U.S. Dollars)

Hardware Revenue 75 72 92

111

Workstation Revenue 0 0 0

0

Peripheral Revenue (Turnkey) 15 16 24

8

Software Revenue 74 127 135

143

Bundled 39 31 34

32

UnbLndled 35 96 101

110

Service Revenue 18 24 39

50 total Revenue 168 222 266

303

Increase over Pri«r Year 105X 33% 20%

14X

228

174

0

54

237

59

179

92

558

18%

TITLE:

APPLICATION:

REGION:

PLATFORM:

43

Final History and Forecast

IC Layout

Worldwide

Host/Server

@

f^

o

E

•g

§

3 a

K

•a

S

I

•o

ti fi g-

1

§•

B.

a

1

1986

=s==

UNIT SHIPMENT DATA (Workstation

Shipments)

CPU Shipments

Workstation Shipments

268

488

CPU Installed Base

Workstation Installed Base

917

2.675

1987

====

198

459

1,029

3,133

1988

3 S = =

1989

===s

510

537

1,420

3,671

925

1,024

2,212

4,694

1990

====

1991

====

1992

====

1993

====

953

968

2,982

5,419

940

940

3,740

6,600

910

910

4,500

7,510

840

840

5,230

8,350

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 463.7

375.5 212.4 601.9

Hardware-Only ASP 193.8

308.3 81.5 44.0

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue

CPU Revenue

63

38

61

33

Workstation Revenue

17 21

Peripheral Revenue (Turnkey) 8

7

Software Revenue

Bmdled

UnbuKtled

26

8

19

24

5

Service Revenue

Total Revenue

12

102

19

10

95

Increase over Prior Year

• U

•7X

44

33

7

4

20

0

20

15

79

-17X

48

41

3

3

19

3

16

14

81

2X

83.5

42.0

42

36

3

3

17

1

16

13

73

-10X

77.9

38.7

36

17

16

3

21

3

18

13

70

-4X

74.6

35.8

32

15

14

3

23

3

20

12

68

-3%

71.5

33.2

27

12

12

3

24

3

22

11

63

•7%

TITLE:

APPLICATION:

REGION:

PLATFORM:

44

Final History and Forecast

IC Layout

Uorlduide

Personal Computer

«

I

I

n

a

Sf

T)

I

g-

3

•fl

3 a

1986 1987 1988 1989 1990

1991 1992

1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 612 1,739 1,420 491

320

Workstation Shipments 612 1,739 1,420 491 320

CPU Installed Base 659 2,396 3,778 4,079 4,052

Workstation Installed Base 659 2,396 3,778 4,079 4,052

400

400

3,470

3,470

470

470

3,110

3,110

350

350

2,630

2,630

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP

Hsrduare-Only ASP

25.7 6.0 15.6 23.5 27.0

2.9 5.0 5.0 3.7 3.1

REVEI4UE DATA (Millions of U.S. Dollars)

Hardware Revenue 2 8 9 4 3

CPU Revenue 2 8 8 3 2

Workstation Revenue 0 0 0 0

0

Peripheral Revenue (Turnkey) 0 0 0 0

0

Software Revenue 4 16 5 2

1 a m l l e d 0 1 1 0

0

Unbundled 4 15 4 2 1

Service Revenue 0 1

Total Revenue 7 25 14 6

4

Increase over Priar Year 121X 266X -45% -54X

-40X

26.3

3.1

3

3

0

0

2

1

1

1

6

51X

26.1

3.0

1

1

1

6

14X

4

3

0

0

2

25.7

3.0

2

2

0

0

1

1

1

1

4

•33X

TITLE:

APPLICATION:

REGION:

PLATFORM:

45

Final History and Forecast

IC Layout

North America

All Platforms

6

«

s*

•g

S

3

8

s

a

!f

•o f i i

n

§.

o'

3

Q* s

a

1986

====

1987

====

1988

====

1989

====

1990

====

1991

====

1992

====

1993

====

CPU Shipments

Workstation Shipments

CPU Installed Base

Workstation Installed Base

1.873

1,989

3,193

4,373

1,880

1,985

4.965

6,321

2,470

2.467

7.179

8,629

3,950

3,956

10,590

12.153

4.657

4.668

15,261

16,969

6,260

6,260

19,090

20,880

8,200

8,200

25,890

27,750

9,810

9,810

34,310

36,220

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S.

Dollars)

Turnkey ASP

90.9 64.8 64.7

Hardware-Only ASP 27.5

25.3 22.3

52.0

17.9

57.4

16.7

54.7

14.4

52.6

12.8

51.3

11.9

REVENUE DATA (Millions of U.S. Doll ars)

Hardware Revenue

CPU Revenue

Uorkstatitjo Revenue

Peripheral Revenue (Turnkey)

Software Revenue

Bundled

Ur^xndLed

Service Revenue

Total Revenue

Increase over Prior Year

70

51

7

12

64

24

40

20

154

50%

52

34

8

10

91

7

83

17

159

3X

59

46

3

10

65

9

57

23

148

•7%

74

69

1

4

71

10

61

32

177

20X

83

77

2

4

75

14

60

43

201

14X

87

63

6

18

76

19

58

41

205

2X

99

74

5

20

80

21

60

44

223

9%

109

82

4

23

89

22

67

47

245

10X

TITLE:

APPLICATION:

REGION:

PLATFORM:

46

Final History and Forecast

IC Layout

North America

Technical Workstation

o

I

I

i

I

s

a

if

•o

!!

3

S"

•V

I

I

I

1986

====

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments

Workstation Shipments

1,126

1,126

2,013

CPU Installed Base

Workstation IrataIled Base 2,013

1987

====

1,226

1,226

3,203

3,203

1988

====

1989

=sss

1990

====

1991

====

1992

====

1993

====

1,631

1,631

4,706

4,706

3,183

3,183

7,577

7,577

3,923

3,923

11,804

11,804

5,540

5,540

15,500

15,500

7,480

7,480

22,040

22,040

9,240

9.240

30,330

30,330

CALCULATED AVERAGE SYSTEM PRICE

DATA (Thousands of U.S. Dollars)

Turnkey ASP

76.0

62.7 62.5

Harduare-Only ASP

20.4 20.7

18.8

51.9

15.0

REVENUE DATA (Hi I lions of U.S.

}oll8rs)

Hardware Revenue

CPU Revenue

UorkstBtion Revenue

37

30

0

30

22

0

Peripheral Revenue (Turnkey) 7

7

Software Revenue

Bundled

46

21

69

7

Unbundled

Service Revenue

Total Revenue

Increase over P r W Year

26

12

95

9^X

63

12

111

16%

34

27

0

8

55

9

46

16

105

•5X

52

49

0

3

59

10

49

25

136

29X

57.6

14.6

64

61

0

3

67

14

53

36

167

23X

55.0

13.3

72

56

0

16

68

18

50

35

175

5X

52.8

12.2

20

53

38

197

13%

86

67

0

19

73

51.4

11.6

99

77

0

22

82

22

60

43

224

14X

TITLE:

APPLICATION:

REGION:

PLATFORM:

47

Final History and Forecast

IC Layout

North America

Host/Server

o

o

1

<n

1

O

1

3

5P

a

I

1

1

1

IT

1

1986 1987

*~~~

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments

Workstation Shipments

152

269

CPU Installed Base

578

Workstation Installed Base 1.757

72

177

579

1.934

1988

= = = S

1989

S = = =

1990

ssss

1991

====

1992

~~~~

1993

~~~~

262

259

743

2,193

494

500

1.131

2,694

480

491

1.586

3,294

430

430

1,820

3,620

390

390

2,150

4.000

310

310

2.410

4.310

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP

Hardware-Only ASP

294.0

200.0

130.3

304.6

183.0

79.1

.0

41.9

94.0

38.5

89.8

35,1

86.1

32.2

82.5

29.5

REVENUE DATA (Hillions of U.S. Dollars)

Hardware Revenue

CPU Revenue

31

19

19

9

Workstation Revenue

7 8

Peripheral Revenue (Turnkey) 5

2

SoftMSre Revenue

Bundled

Unbundled

15

4

12

11

1

11

Service Revenue

Total Revenue

Increase over Prior Year

7

54

4X

4

34

-37%

22

16

3

2

9

0

8

7

37

8%

21

18

1

1

11

0

11

7

39

4X

19

15

2

1

7

0

7

7

32

-16X

15

7

6

2

7

1

7

6

28

-15%

12

5

5

2

7

1

6

5

24

-13%

9

4

4

1

6

0

6

4

19

-19%

TITLE:

APPLICATION:

REGION:

PLATFORM:

48

Final History and Forecast

IC Layout

North America

Personal Computer

s

D u

a

Ul

§

•g

S a

X"

3

S"

n

c

§•

3

•fl

3 g;

s?'

a

1986

====

UNIT SHIPHEHT DATA (Workstation Shipments)

CPU Shipments

594

Workstation Shifments

594

CPU Installed Base

Workstation Installed Base

602

602

1987

====

581

581

1,183

1,183

1988

====

577

577

1,729

1.729

1989

====

273

273

1,882

1,882

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP

Hardware-Only ASP

25.8

2.9

32.7

5.0

25.8

5.3

49.0

3.5

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue

CPU Revenue

Workstation Revenue

Peripheral Revenue (Turnkey)

Software Revenue

Bmdled

Unbundled

Service Revenue total Revenue

Increase over Pric* Year

3

0

3

0

5

2

2

0

0

298X

3

3

0

0

10

0

10

1

14

190%

3

3

0

0

2

0

2

0

5

-62X

1

0

1

0

2

-59X

1

1

0

0

1990

====

254

254

1,871

1,871

24.7

3.1

1

0

1

0

2

-17X

1

1

0

0

1991

====

290

290

1,760

1,760

24.5

3.1

1

0

1

0

2

18%

1

1

0

0

1992

====

330

330

1,700

1,700

24.2

3.0

1

0

1

0

2

3X

1

1

0

0

1993

s===

260

260

1,570

1,570

24.0

3.0

1

0

1

0

2

-26X

1

1

0

0

TITLE:

APPLICATION:

REGION:

PLATFORM:

49

Final History and Forecast

IC Layout

Europe

All Platforms

S

•9

5"

8

i

s

& if

"O f?

3

•o

9

n

I

§•

CT

I

1986 1987 1988 1989 1990

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 104.2 30.1 56.8 54.3 53.5

HarduareOnly ASP 46.7 25.5 24.6 20.7 21.1

1991

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 377 1,277

1,491

Workstation Shipments 390 1,304

1.498

CPU Installed Base 683 1,950 3,404

Workstation Installed Base 775 2,070 3,534

1,865

1,839

5,131

5,243

2,238

2,227

7,396

7,505

2,990

2,990

9.380

9,530

3,870

3,870

12,620

12.790

4,830

4,830

16,840

17,030

51.0

18.5

1992

48.9

16.4

1993

47.7

14.9

REVENUE DATA (Hillions of U.S. Dollars)

Hardware ReverKie 20 29

40

CPU Revenue 16 21

33

Workstation Revenue 1 3 2

Peripheral fteverwe (Turnkey) 3 5 5

Software Revenue 19 37 32

Bundled 6 5

6

Ur^undled 13 32

27

Total Revenue 42 76

87

Increase over Prt«r* Year 25X BOX

15X

41

38

1

2

26

4

22

14

81

-6X

49

46

1

2

29

6

23

21

99

22X

56

40

5

11

31

8

23

23

111

12X

64

46

5

13

35

9

26

26

125

13X

73

53

5

15

41

9

32

30

144

15X

TITLE:

APPLICATION:

REGION:

PLATFORM:

50

Final History and Forecast

IC Layout

Europe

Technical Workstation

"8

3

Z

a

3

I

(I

•o

I

I

^•f

1

s

at a

!f

•o fii

3

!f

«

R

1986 1987 1988 1989 1990 1991

1992

1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments

314

Workstation Shipments

CPU Installed Base

Workstation Installed Base

314

529

529

439

439

961

961

842

842

1,774

1,774

1,543

1,543

3,237

3,237

1,978

1,978

5,272

5,272

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousarxis of U.S.

Dollars)

92.3

21.4

77.6

21.5

63.3

21.4

56.0

15.6

53.6

16.9

2,710

2,710

7,430

7,430

3,570

3,570

10,690

10,690

4,540

4,540

14,910

14,910

51.1

15.3

49.0

14.0

47.7

13.3

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue

11

CPU Revenue

9

Workstation Revenue

0

Peripheral Revenue (Turnkey)

Software Revenue

Suidled

2

11

5

Unbundled

Service Revenue

6

2

Total Revenue

Increase over Prior Year

25

58X

13

9

0

4

29

5

24

6

48

94X

22

17

0

4

28

5

23

9

59

23X

27

26

0

1

25

4

21

11

63

8X

36

35

0

1

27

6

21

18

81

28X

44

35

0

10

28

7

21

20

92

14X

53

41

0

11

32

7

24

22

107

16X

62

49

0

14

39

8

30

26

127

19X

TITLE:

APPLICATION:

REGION:

PLATFORM:

51

Final History and Forecast

IC Layout

Europe

Host/Server

©

>-*

o n

•§

S f

1

s

if

"O ft s==s

UNIT 5H1PHENT DATA (Uorkstation

Shipments)

CPU Shipments

Uorkstation Shipments

CPU Installed Base

48

61

104

Uorkstation Installed Base

1986

196

1987

===a

46

74

150

270

I c

"O

5 c.

3

t

1

1

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue

CPU Revenue

Uorkstation Revenue

9

6

1

13

9

3

Peripheral Revenue (Turnkey) 1 1

Software Revenue

Bmdled

Unbundled

6

1

5

7

0

7

Service Revenue 1

16

3

23

Total Revenue

Increase over Prior Year -3% 41X

1988

==r=:

169

177

317

447

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

236.1 244.7

Turnkey ASP

Hardware-Only ASP

355.5

167.9 299.6 89.0

.0

53.3

16

13

2

1

3

0

3

6

25

10%

1989

====

244

217

553

665

13

12

1

1

0

0

0

3

16

-35%

.0

49.7

1990

====

1991

====

260

249

762

871

280

280

1,050

1,190

1992

===s

1993

====

300

300

1.320

1,490

300

300

1,590

1,790

13

11

1

1

2

0

2

4

19

14X

.0

46.4

12

5

5

1

3

1

2

4

18

-0%

.0

42.7

11

5

5

1

3

1

2

4

18

-4X

.0

39.2

10

5

5

1

3

1

2

3

16

-8X

TITLE:

APPLICATION:

REGION:

PLATFORM:

52

Final History and Forecast

IC Layout

Europe

Personal Computer

@

»-»

•§

§

}f

•o f?

3

?r rt

•O

I

§•

9

•O

3 q-

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 15 792 480 78 0 0 0 0

Workstation Shipments 15 792 480 78 0 0 0 0

CPU Installed 8ase 49 839 1.313 1,341 1.362 900 620

330

Workstation Installed Base 49 839 1.313 1,341 1.362 900 620 330

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 19.6 2.8 6.5 33.0 .0 .0 .0

Hardware-Only ASP 2.9 5.0 4.7 3.7 .0 .0 .0

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 0 3 2

CPU Revenue 0 3 2

1

1

0 0 0

0

0 0 0

0

Workstation Revenue 0 0 0 0 0 0 0

0

Peripheral Revenue (Turnkey) 0 0 0 0 0 0 0

0

Software Revenue 1 2 1 1 0 0 0

0

Bundled 0 0 0 0 0 0 0

0

Unbundled 1 2 1 1 0 0 0

0

Service Revenue 0 0 0 0 0 0 0

0

Total Revenue 1 5 3 2 0 0 0

0

TITLE:

APPLICATION:

REGION:

PLATFORM:

53

Final History artd Forecast

IC Layout

Asia

All Platforms

•-»

O

»

B

§

•g

a

I

}p

•o

Vi*

I

ft g-

9

P-

I

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPHENT DATA (Workstation Shipments)

CPU Shipments 394 1.096 1,270 1,693

2,263 3,260

4,420

Workstation Shipments 480 1,224 1.290 1,809 2,277 3,260

4.420

CPU Installed Base 767 1,855 3,092 4,662 13,150

Uorkstation Installed Base 1,135 2,353 3,614 5,317 6.959 10,060 13,960

5,730

5,730

18,420

19.270

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turr>key ASP 229.1 148.2 108.4 88.4 62.0 57.3

54.0

Hardware-Only ASP 63.1 29.4 15.3 13.6 14.6 12.7

11.4

54.0

10.7

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 45 57 43 45 57 58 65 74

CPU Revenue 29 40 29 38 50 39 45 51

Workstation Revenue 8 10 1 1 0 4 4 3

Peripheral Revenue (Turnkey) 8 8 13 6 7 15 17 20

Software Revenue 21 38 62 67 78 89 107 131

Bundled 16 23 21 21 18 26 28 30

Unbundled 5 14 41 46 61 63 80 101

Service Revenue 6 9 16 17 16 18 22 26

Total Revenue 72 104 121 129 151 165 195 231

Increase over Prior Year 47X 44X 16% 6% 18X 9% 18% 19%

TITLE:

APPLICATION:

REGION:

PLATFORM:

5A

Final History and Forecast

IC Layout

Asia

Technical Workstation

s*

B

1

a

a

Jf

T3

?f

3

n

•o

I

c

§•

9

T)

I a

I

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 343 714 863 1.388 2.008 2.950 4,080 5,440

Workstation Shipments 343 714 863 1,388 2,008 2,950 4,080 5,440

CPU Installed Base 572 1,279 2,114 3,414 5,191 7,780 11,480 16,640

Workstation Installed Base 572 1,279 2.114 3.414 5,191 7.780 11.480 16.640

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 171.2 129.6 148.1 82.9 62.4 59.5 57.1 55.7

Hardware-Only ASP 25.7 24.0 12.7 10.7 12.9 11.7 10.7 10.2

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 25 29 35 30 45 47 55 65

CPU Revenue 20 24 23 26 39 33 39 46

Workstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 6 4 12 4 6 15 16 19

Software Revenue 16 29 52 58 70 77 93 115

Bundled 13 19 21 18 16 23 25 28

Unbundled 3 10 32 41 54 54 68 87

Service Revenue 4 6 14 13 14 15 18 22 total Revenue 45 63 101 102 129 140 166 203

Increase over PriQT Year 188X 41X 60X OX 27X 8X 19% 22X

TITLE:

APPLICATION:

REGION:

PLATFORM:

55

Final History and Forecast

IC Layout

Asia

Host/Server

o

»

•g

S

o

3

if

TD

?B

1

I

•o

1 c. o

9

?

g:

cr

1

1986

====

UNIT SHIPMENT DATA (Uorlcstation

Shipments)

CPU Shipments

Workstation Shipments

48

134

CPU Installed Bsse

Workstation Installed Base

188

556

1987

====

71

199

258

755

1988

====

68

89

322

844

1989

====

169

285

474

1,129

CALCULATED AVERAGE SYSTEM PRICE

DATA (Thousands of U.S.

Dollars)

Turnkey ASP 1223.1

478.2 216.0

598.6

Hardware-Only ASP

213.3 353.8 65.9 34.2

1990

==r=

190

204

562

1.022

83.6

37.8

1991

====

1992

r===

200

200

770

1,530

190

190

920

1,730

78.2

33.0

75.0

30.3

1993

====

200

200

1,070

1,930

71.8

27.9

Hardware Revenue

CPU Revenue

Workstation Revenue

Software Revenue

Bundled

Ufixnlled

19

10

8

5

3

2

27

14

10

Peripheral Revenue (Turnkey) 2

3

5

4

1

Service Revenue

Total Revenue

3

27

3

35

Increase over Pri«r Year

-20X

32X

5

3

1

0

8

0

8

2

15

-58%

13

10

1

2

8

3

5

3

25

66X

10

9

0

1

8

1

6

3

20

^^n

9

4

4

0

11

1

9

3

22

10%

8

4

4

0

13

1

12

3

24

9X

7

3

3

0

15

1

14

3

26

6X

TITLE:

APPLICATION:

REGION:

PLATFORM:

56

Final History and Forecast

IC Layout

Asia

Personal Coniputer

®

i n

O.

if

•D

K

i

I

n

TD

I

i

3

3 cr

I

1986

1987

1988

1989

UNIT SHIPMENT DATA (Uorkstation Shipments)

CPU Shipments 3 311 339

135

Workstation Shipments 3 311 339

135

CPU Installed Base 7 318 656

774

Uorkstation Installed Base 7 318 656

774

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

26.3 15.1 18.3 20.1

.0 5.0 5.0

4.2

1990

65

65

746

746

27.2

.0

1991

110

110

750

750

26.4

.0

1992

140

140

750

750

1993

90

90

700

700

26.1

.0

25.8

.0

Hardware Revervue

CPU ReverKie

Uorkstation Reveruje

Peripheral Revenue (Turnkey)

Software Revenue

Bundled

Unbundled

Service Revenue

Total Revenue

Increase over Prior Year

0

0

0

0

1

0

1

0

1

138X

2

2

0

0

4

1

3

0

6

65n

3

3

0

0

2

1

1

0

5

•12X

2

2

0

0

0

0

0

0

2

-57X

2

2

0

0

0

0

0

0

2

-10X

2

2

0

0

1

1

0

0

3

69%

3

2

0

0

1

1

0

0

4

23%

2

1

0

0

1

1

0

0

3

-37X

TITLE:

APPLICATION:

REGION:

PLATFORM:

57

Final History and Forecast

IC Layout

Rest of Uorld

All Platforms

©

^rt

»-*

D

•s

8

1

a

I

•o

H

1

I

n

•o

I

c

C3

1

====

UNIT SKIPHENT DATA (Workstation

Shipments)

CPU Shipments

Uorkstation Shipments

CPU Installed Base

Workstation trKtalled Base

1986

75

79

121

239

1987

====

74

73

180

312

1988

s===

67

70

231

380

1989

s===

119

122

339

492

CALCULATED AVERAGE SYSTEM PRICE

DATA (Thousands of U.S.

Dollars)

Turnkey ASP

116.8 60.7

86.8

Hardware-Only ASP 69.5 34.7 29.4

47.9

19.9

REVENUE DATA (Hi I lions of U.S. Dollars)

Hardware Revenue

CPU Revenue

Workstation Revenue

5

4

1

3

2

0

Peripheral Reverxie (Turnkey) 1

0

Softuare Revenue

Bundled

Unbirdted

Service Revenue

Total Revenue

1

1

0

1

7

0

0

0

0

3

Irtcrease over Prior Year 262X -55X

2

2

0

0

0

0

0

1

3

-5X

0

1

1

4

23X

3

2

0

0

1

1990

====

122

124

431

590

31.7

19.4

0

0

1

3

-21%

2

2

0

0

0

1991

====

150

150

530

690

27.8

17.6

0

0

1

4

16X

2

2

0

0

1

23.5

15.9

1992

==r=

170

170

650

820

0

1

1

4

18X

2

2

0

0

1

1993

====

200

200

800

980

30.3

14.6

0

1

1

5

25X

3

2

0

0

2

TITLE:

APPLICATION:

REGION:

PLATFORM:

58

Final History and Forecast

IC Layout

Rest of World

Technical Uorkstation

®

I

n

T3

•O

3

g-

3

I

•g

a

H

a

if

•o fS

3

!f

1986 1987 1988

1989

UNIT SHIPMENT DATA (Uorkstation Shipments)

CPU Shipments 56 10 32

98

Uorkstation Shipments 56 10 32

98

CPU Installed Base 74 83 113

203

Workstation Installed Base 74 83 113

203

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

80.4 53.9 77.4

24.9 28.6 18.2

39.6

13.4

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue

'2

0 1

1

CPU Revenue 2 0 1

1

Uorkstation Revenue 0 0 0

0

Peripheral Revenue (Turnkey) 0 0 0

0

Software Revenue 1

Bundled 0 0 0

0

Urixjndled 0 0 0

1

Service Revenue 0 0 0

0

Total Revenue 3 1 1

2

Increase over Prior Year 274% -78X 85%

110X

1990

100

100

287

287

35.1

13.4

0

0

0

1

1

0

0

0

2

-27%

1991

120

120

380

380

33.5

12.2

0

0

0

1

1

0

0

0

2

19%

1992

140

140

490

490

32.1

11.2

1

1

0

0

1

0

1

0

3

26X

1993

170

170

640

640

31.3

10.6

2

1

0

0

1

0

1

1

4

38X

TITLE:

APPLICATION:

REGION:

PLATFORM:

59

Final History and Forecast

IC Layout

Rest of World

Host/Server

&

I—»

•S

§

•i

if

TT

3

5f

cr

3

I

§•

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Uorlcstation Shipments)

CPU Shipments 19 9 10 17 22 30 30 30

Workstation Shipments 24 8 13 21 24 30 30 30

CPU Installed Base 48 42 37 54 72 100 120 150

Workstation Installed Base 165 173 186 207 232 260 290 320

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 317.9 187.5 208.3 529.4 .0 .0 .0 .0

Hardware-Only ASP 178.1 237.2 122.6 57.3 46.2 42.1 38.6 35.4

REVENUE DATA (Hi I Lions of U.S. Dollars)

Hardware Revenue 4 2

CPU Revenue 3 1 1

Workstation Revenue

Peripheral Revenue (Turnkey)

Software Revenue

Smdled l

O

O

O

O

O

O

O

O

O

O

O

1

1

1

O

O

O

O

1

O

O

O

O

1

0 0 0

O

O

O

O

1

O

O

O

O

1 1

O

O

O

O

Unbundled

Service Revenue l

O O

O

Total Revenue 5 2 2

O

O

1

O

O

1

O

O

1

O

O

O

O

O

O

2 2

Increase over Prior Year 255% -50X -22% -25% -4% 1 U 5% ,1%

TITLE:

APPLICATION:

REGION:

PLATFORM:

60

Final History and Forecast

IC Layout

Rest of World

Personal Conputer

@

D

K

B

I

I

!f

•o

3

S"

T3 g-

I

9

TJ

3 cr

1986 1987 1988

1989

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 0 55

25

Workstation Shipments 0 55

25

CPU Installed Base 0 55

80

Workstation Installed Base 0 55

80

4

4

82

82

1990 1991

0

0

72

72

0

0

50

50

1992

0

0

40

40

1993

0

0

20

20

Turnkey ASP .0 .0

.0

Hardware-Only ASP .0 4.9

>.0

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 0 0

0

CPU Revenue 0 0

0

Workstat ton Revenue 0 0

0

Pertptieral Revenue (Turnkey) 0 0

0

Software Revenue 0 0

0

Bundled 0 0

0

Unbundled 0 0

0

Service Revenue 0 0

0

Total Revenue 0 0

0

Increase over Prior Year MA NA

«9X

25.1

5.5

0

0

0

0

0

0

0

0

0

-44%

.0

.0

0

0

0

0

0

0

0

0

0

-100X

.0

.0

0

0

0

0

0

0

0

0

0

NA

0

0

0

0

0

0

0

0

0

OX

0

0

0

0

OX

0

0

0

0

0

TITLE:

APPLICATION:

REGION:

PLATFORM:

61

Final History and Forecast

PCB/Hybrid/MCM

Worldwide

All Platforms

6

o

g

§

I

3

if

•u

S

I

c

n

R

0

1

9

?

E

ff

1

1986 1987 1988 1989 1990 1991

1992 1993

S S = =

====

==== ====

=s=s s===

====

UNIT SHIPMENT DATA (Workstation Shipments)

Workstation Installed Base

29,992

23,280

25,372

25,517

27,633

28,800

30,830

32,480

34,260

36,120

37,590

4

4

86,266 106,840 118.390

137,050 159,610

18

24.261 46,991 75,765 97,573 120,356 134,710 155,630 180,080 20

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 115.5 74.3 80.6

Hardware-Only ASP

16.0 11.2

9.6

55.1

15.9

REVENUE DATA (HiUions of U.S. Dollars)

Hardware Revenue

399

CPU Revenue

274

Workstation Revenue

Peripheral Revenue (Turnkey)

Software Revenue

69

56 fiundted

Urtundled

Service Revenue

Total Revenue

Increase over Prior Year

179

126

53

71

649

38%

430

294

70

67

270

125

145

113

813

25X

511

386

40

85

296

151

145

128

934

1SX

518

411

45

63

305

182

122

136

958

3X

57.4

14.7

535

429

43

63

326

180

146

164

1,026

7%

57.1

13.9

564

380

58

125

332

178

154

176

1,073

5%

55.6

13.1

586

398

56

132

356

184

172

189

1,131

5%

54.5

12.8

600

413

50

137

384

181

204

201

1,186

5X

TITLE:

APPLICATION:

REGION:

PLATFORM:

62

Final History and Forecast

PCB/Hybrid/MCM

Worldwide

Technical Workstation

@

I - '

•o

I

c.

§

•fl

3 g-

i n a.

}f

•o fl

3

K

1986 1987

1988 1989 1990 1991 1992

1993

UNIT SHIPHENT DATA (Workstation Shipments)

CPU Shipments 3,168 3,891 6,143 10,964

11,999 14,330 18,220

Workstation Shiponnts 3,168 3,891

6,143 10,964

11,999 14,330

18,220

CPU Installed ease 5,548 9,339 15,136 25,245

39,230 47,010 62,400

39,230 47,010 62,400

24,410

24,410

84,070

84,070

1

1

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 95.8 93.8 79.1 61.9

62.6

Hardware-Only ASP 24.1 22.8 20.2 15.0 15.2

59.0

13.8

56.4

12.6

54.8

11.9

REVENUE DATA (Hi I lions of U.S. Dollars)

Hardware Revenue 163 179 233 288 305 341 378 427

CPU Revenue 136 138 178 240 260 246 274 313

Workstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 26 41 56 47 45 95 104 115

Software Revenue 89 154 193 234 259 259 290 328

Bundled 81 88 111 160 156 149 160 162

Unbundled 8 66 82 74 103 110 130 166

Service Revenue 27 60 78 93 122 134 150 168

Total Revenue 279 393 504 615 687 734 818 924

Increase over Prior Year 62% 41% 28% 22% 12%

7%

11% 13%

TITLE:

APPLICATION:

REGION:

PLATFORM:

63

Final History and Forecast

PCB/Hybrid/MCM

Worldwide

Host/Server

3

0 s

@ i

D u

C

•s u if i,

•o

1

§

1986

====

IMIT SHIPHENT DATA (UorksMtion Sh ipments)

594 CPU Shipments

Workstation Shipments

CPU Installed Base

Workstation Installed Bise

2,083

2,268

8,027

1987

====

575

2,241

2,603

10,269

1988

====

881

1,883

3,194

12,152

1989

====

1,171

3.264

4,108

15,416

1990

====

1,294

3,410

4,913

18,429

1991

====

1,430

3,460

5,980

22,290

1992

====

1,440

3,230

6,930

25,520

1993

====

1,340

2,810

7,860

28.320

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S

Turnkey ASP

462.5 336.0

Hsrduare-Only ASP 248.5 350.1

Dollars)

460.2

107.4

437.5

110.8

REVENUE DATA (HilUons of U.S. Dollars)

Hardware Revenue 176

CPU Revenue 82

Workstation Revenue

Peripheral Revenue (Turnkey)

Software Revenue

Buxlled

69

26

50

39

10

UnbivKlled

Service Revenue

Total Revenue

Increase over Prior Year

40

266

-1%

168

77

70

21

47

29

18

44

258

-1%

155

96

40

19

46

36

10

41

242

-6%

157

102

45

11

27

20

7

36

220

-9%

164.3

118.0

151

98

43

9

23

19

4

35

208

-5%

158.1

107.1

149

67

58

23

25

22

3

35

209

0%

150.1

98.9

137

61

56

21

21

19

2

33

191

-8%

146.8

90.8

118

51

50

17

16

15

2

28

162

-15%

TITLE:

APPLICATION:

REGION:

PLATFORM:

64

Final History and Forecast

PCB/Hybrid/MCM

Worldwide

Personal Computer

3

n

a

ff

•o

?J

3

•o

I n g-

3

•fl

3

ST. ai

9

o

1986 1987

1988 1989

1990 1991 1992

1993

UNIT SHIPMENT DATA (Workstitfon Shipments)

CPU Shipments 8,172 16,906 21,966

Uorkstation Shipments 8,172 16,906

21,966

CPU InstaUed esse 10,685 27,383 48,478

Uorkstation Installed Base 10,685 27,383 48,478

11.145

11.145

56,913

56,913

12,224

12,224

62,697

62,697

13,040

13,040

65,410

65,410

12,810

12,810

67,720

67,720

10,370

10,370

67,680

67,680

CALCULATED AVERAGE STSTEH PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

24.0 11.0 10.7

6.5 4.6 5.4

12.1

5.4

16.2

5.4

16.5

5.3

16.4

5.2

16.1

5.2

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 60 84 123 73 79 74 70 55

CPU Revenue 56 79 112 69 71 67 63 49

Uorkstation Revenue

Peripheral Revenue (Turnkey) 4

0 0 0

4 11

0

4

0

8

0

7

0

7

0

6

Software Revenue 40 69 57 44 44 49 45 40

Buidled 5 8 4 3 5 7 6 4

UnbuidLed 35 62 53 41 40 42 39 36

Service Revenue 4 9 9 7 7 7 6 5

Total Revenue 104 162 189 123 130 130 121 100

Increase over Prior Tear 248X 56X 16X -35% 6X -IX -7X -17X

TITLE:

APPLICATION:

REGION:

PLATFORM:

65

Final History and Forecast

PCB/Hybrid/MCM

North America

All Platforms

•§

R o

%

3

I

if

•o

•8

I r>

I

§•

o

I

a

I

1986

1987 1988

1989

1990

1991

1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

Workstation Shipments 6,222

Workstation Installed Base

14,776

31,966

9,414

10,189

39,369

10,175

11,145

48,056

12,050

12,940

50,870

13,720 16,210

14,420 16,720

57,990 67,630

12,727 22,268 36,360 44,731 54,863 58,550 66,600 76,950

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP

Harduare-Only ASP

96.3 75.6 89.4 61.0 61.6 57.1 53.3 51.2

14.4 9.3 8.3 15.0 14.8 13.6 12.5 12.1

REVENUE DATA (Millions of U.S. Dollars)

HardMare Revenue 148 128 176 168 181 198 202 212

CPU Revenue 107 88 144 136 145 140 146 156

Workstation Revenue 23 24 9 18 22 22 18 14

Peripheral Revenue (Turnkey) 18 16 22 14 14 36 39 42

Software Revenue 58 77 94 85 98 100 107 113

Bundled 29 22 36 38 41 42 44 40

Unbundled 29 55 58 47 57 58 63 73

Service Revenue 31 40 45 50 67 73 76 79

Total Revenue 238 246 315 303 346 371 385 403

Increase over Prfbr Year 9X 3X 28X -4X 14X 7X 4X 5X

TITLE:

APPLICATION:

REGION:

PLATFORM:

66

Final History and Forecast

PCB/Hybrid/HCM

North America

Technical Workstation

©

t - *

t

3 fj a if

•o

3

I n

•O

§

I

1986 1987 1988 1989

79

68

0

11

59

35

24

33

171

8X

1990

88

78

0

10

76

37

39

50

214

25X

1991

108

82

0

26

80

38

42

57

245

14X

1992

126

96

0

29

90

41

48

63

278

14X

1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 1,389 1,367 2,322

Workstation Shipments 1,389 1,367 2,322

CPU Installed Base 2,802 4,108

6,241

Workstation InstsUed Base 2,802 4,108 6,241

3,882

3,882

9,679

9,679

4,465 5,730

7,760

4,465 5,730 7,760

15,679 17,830 24,360

15,679 17,830 24,360

11,480

11,480

34,620

34,620

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP

74.6 74.4 76.5 54.9

56.8

Hardware-Only ASP 22.8 21.7

20.3 15.4

15.0

54.2

13.6

52.0

12.5

50.7

11.9

REVENUE DATA (Hi 11 ions of U.S. Dollars)

Hardware Sevenue 58 48

75

CPU Hevenue 48 37

57

Workstation Revenue 0 0

0

Peripheral Revenue (Turnkey) 10 11

18

Software Revenue 26 34

57

Bundled 21 17

31

Unbundled 5 17

25

Service Revenue 12 19

27

Total Revenue 96 101

158

57%

Irtcrease over Prior Year 23X 5X

153

118

0

35

99

39

60

69

320

15X

TITLE:

APPLICATION:

REGION:

PLATFORM:

67

Final History and Forecast

PCB/Hybrid/MCM

North America

Host/Server

o

*-*

D

•s

S

1

1

a

I

•o

S

1

I

•3

1

1

S

H. o-

1

1

1986

====

UNIT SHIPMENT DATA (Workstation

Shipments)

CPU Shipments

240

Uorkstation Shipments

710

CPU Installed Base

1,116

Workstation Installed Base 4,301

1987

===s

175

916

1,178

5,216

1988

===s

1989

===s

364

558

1,380

5,774

565

1,340

1,752

7,113

1990

====

1991

====

1992

====

1993

====

510

1,480

2,153

8,961

550

1,440

2,350

10,030

490

1,180

2,610

11,220

420

920

2,820

12,140

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP

443.1 353.1 284.5 302.0

Harduare-Only ASP 197.1

280.8 104.7 103.8

387.9

116.4

370.5

106.0

355.3

97.3

340.5

89.2

REVENUE DATA (Hi I Mora of U.S. Dollars)

Hardware Revenue

CPU Revenue

Workstation Revenue

63

32

23

50

20

24

Peripheral Revenue (Turnkey

) 8

5

Software Revenue

BLndled

Unbundled

Service Revenue

Total Revenue

15

7

7

17

95

13

4

9

18

81

Increase over Prior Year -22% -15%

45

32

9

4

8

4

4

14

68

•16%

63

42

18

3

6

3

3

15

84

24%

65

40

22

3

5

4

2

15

85

1%

60

29

22

10

4

3

1

14

78

-A

49

23

18

8

3

2

1

11

62

-21%

37

17

14

6

1

1

0

8

47

-25%

TITLE:

APPLICATION:

REGION:

PLATFORM:

68

Final History arxi Forecast

PCB/Hybrid/MCM

North America

Personal Computer

©

t-t

I

§

3 a

if

T3

H

B n

•V

I

c

§•

3

•O

3 a

CPU Shipments

Workstation Shipments

CPU Installed Base

Workstation Installed Base

1986

====

1987

====

1988

====

1989

====

1990

====

1991

====

1992

====

1993

====

4,124

4,124

5.624

5,624

7.456

7,456

12,944

12,944

11.896

11,896

24,345

24,345

4,968

4,968

27,939

27,939

5,200

5,200

30.224

30.224

5,780

5,780

30,680

30,680

5,470

5.470

31.010

31,010

4,320

4,320

30,190

30,190

Turnkey ASP

Hardware-Only ASP

17.5

6.5

11.9

4,1

320.0

4.5

12.9

5.3

17.1

5.2

16.9

5.1

16.8

5.1

16.6

5.0

Hardware Revenue

CPU Revenue

Uorkstscion Revenue

Peripheral Revenue (Turnkey)

Software Revenue

Bundled

Unbundled

Service Revenue

Total Revenue

Increase over Prior Year

28

27

0

0

17

1

16

2

46

150X

31

31

0

0

30

1

29

3

64

39%

55

55

0

0

29

1

28

4

89

38%

26

26

0

0

19

0

19

2

47

-46%

28

27

0

1

16

0

16

2

46

-2%

30

29

0

1

16

0

15

2

48

3%

28

27

0

1

14

0

14

2

45

-7%

22

21

0

1

13

0

12

2

36

-19%

TITLE:

APPLICATION:

REGION:

PLATFORM:

69

Final History and Forecast

PCB/Hybrid/HCM

Europe

All Platforms

B:

IT

S

i

3 a

w

i

^ •

-o fi

50

n

•o

I

I

9

1986

UNIT SHIPMENT DATA <Uorkst«ion Shipments)

1987 1988 1989 1990 1991

1992 1993

8.839

22.738

7,438

8,190

27,654

29,788

8,496

9.102

34.334

37,151

9,560

10,160

38,470

41,940

10,530

11.100

44.490

48,610

10,340

10,820

50,380

55,030

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 99.6 44.7 48.0 46.8

50.1

Hardware-Only ASP 11.7 10.6 11.5 17.5 14.7

52.2

14.2

51.7

13.9

51.3

14.4

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 123 123 153 160 161 185 198 201

CPU Revenue 101 92 133 138 140 126 135 136

Uorkstation Reveme 9 17 8 17 14 22 23 21

Peripheral Revenue (Turnkey) 13 14 13 5 7 37 41 44

Software Revenue 68 114 96 94 98 99 104 113

Bundled 50 38 34 44 43 42 47 51

Unbundled 17 76 62 50 55 57 58 62

Service Revenue 22 43 45 52 62 69 75 82

Total Revenue 212 280 294 306 322 353 378 396

Increase over Prior Year 140X 32% 5% 4X 5X 10X 7X 5X

TITLE:

APPLICATION:

REGION:

PLATFORM:

70

Final History and Forecast

PCB/Hybrid/MCM

Europe

Technical Workstation

6

I—»

S

a.

•o

K

9

§

I

"TJ

3

E g-

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 1,280 1,342 2,031 3,130 3,440 4,160

5,040 6,060

Workstation Shipments 1,280 1,342 2,031 3,130 3,440 4,160

5,040 6,060

CPU Installed Base 1,946 3,254 5,161 8,001 11,838 14,050

18,150 23,280

Workstation Installed Base 1,946 3,254 5,161 8,001 11,838 14,050

18,150 23,280

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 101.7 72.2 64.2 60.9 54.8 52.4 50.3 49.1

Hardware-Only ASP 27.1 23.3 24.0 16.0 16.8 15.2 13.9 13.2

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 74 51 75 83 82 103 117 133

CPU Revenue 65 43 67 82 82 76 . 86 97

Workstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 9 9 9 1 0 27 31 36

Softusre Revenue 46 72 68 70 73 68 77 89

Bundled 45 32 28 40 38 33 38 44

Unb*#idled 1 40 40 30 36 35 38 45

Service Revenue 10 24 29 37 47 52 59 68

Total Revenue 129 147 172 190 202 224 253 290

Increase over Prior Year 175% 14% 17% 11% 6% 11% 13% 15%

TITLE:

APPLICATION:

REGION:

PLATFORM:

71

Final History and Forecast

PCB/Hybrid/MCM

Europe

Host/Server

R

o

3 a jf

•o ft

«

I

I

I

I-

3

IT

i

I

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPHEHT DATA (Uorkstation Shipments)

CPU Shtpnents 60 118 264 368 352 440 500 500

Workstation Shipments 357 446 472 1,120 957 1,040 1,080 980

CPU InstaUed Base 276 392 647 983 1,211 1,640 2,060 2,510

Workstation Installed Base 1,078 1,524 1,995 3,116 4,029 5,110 6,180 7,160

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP 514.3 358.7 425.7 376.8 380.4 359.2 344.8 329.7

HardwareOnly ASP 513.9 402.0 110.5 133.5 122.9 110.8 102.5 94.3

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 26 41 40 54 49 52 53 47

CPU Revenue 13 21 29 35 33 23 23 20

Workstation Revenue 9 17 8 17 14 22 23 21

Peripheral Revenue (Turnkey) 4 4 3 3 2 7 7 6

Software Revenue 6 11 7 6 5 8 8 7

Bundled 5 4 4 3 4 7 7 6

Ur^xjndled 2 7 3 2 1 1 1 0

Service Revenue 11 14 12 12 13 14 14 13

Total Revenue 44 67 59 72 67 74 75 67

Increase over Prior Year 15X 54X -12X 23X -8X 11X IX -11X

TITLE:

APPLICATION:

REGION:

PLATFORM:

72

Final History and Forecast

PCB/Hybrid/MCM

Europe

Personal Computer

s*

B

1

5P

•o

n

3

•O

I

I

t

"O

3 s: a

1986 1987

1988 1989 1990 1991

1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 2,977 6,1U 6,337 3,940 4,705 4,960 4,990

Workstation Shipments 2,977 6,114

6,337 3,940

4,705 4,960 4,990

CPU Installed Base 3,367 9,456 15,581 18,671 21,285 22,780 24,280

Workstation Installed Base 3,367 9,456 15,581 18,671 21,285 22,780 24,280

3,780

3,780

24,580

24,580

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

13.5 7.1 7.9

7.2

7.4 4.7 5.9 5.6

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 23 30 38

23

CPU Revenue 22 28 37

21

Workstation Revenue 0 0 0

0

Peripheral Revenue (Turnkey) 0 2 1

1

Software Revenue 16 31 21

18

Bundled 1 2 2

1

Unbundled 15 29 19

17

Service Revenue 1

Total Revenue 39 65 63

43

14.8

5.5

31

26

0

5

20

1

19

2

53

21X

16.8

5.5

30

27

0

3

22

2

21

2

54

4X

16.9

5.4

29

26

0

3

20

1

19

2

50

-8X

17.0

5.4

21

19

0

2

17

0

17

2

40

•22X

TITLE:

APPLICATION:

REGION:

PLATFORM:

73

Final History and Forecast

PCB/Hybrid/MCM

Asia

All Platforms

o

»

i

s

I c

n g-

3

I

if

•o f»

3

S"

n

T3 s:

a

1986 1987 1988

1989

1990 1991 1992 1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 1.640 4,424 5,482

Workstation Shipments 2,332 4,989

6,061

6,042

6,534

17,949

20,912

6,417

6,856

22,791

25,700

6,630

7,020

27,140

31,190

7,440

7,770

32,090

36,640

8,470

8,750

38,240

43,230

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S. Dollars)

Turnkey ASP

Hardware-Only ASP

175.1 118.9

118.2

32.1 16.8 10.6

57.8

15.3

59.7

13.8

60.0

12.9

59.1

12.2

58.2

11.7

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 116 172

176

CPU Revenue 58 109 104

Workstation Revenue 34 27 22

Peripheral Revenue (Turnkey) 25 36 49

Bundled 45 64 80

Unbundled 6 13 24

Service Revenue 17 28 37

Total Revenue 184 278

317

Increase over Prtpp Year 16% 51X

14%

182

131

8

43

124

99

25

32

338

7X

184

137

6

41

129

95

33

33

346

2%

170

108

11

51

130

92

38

32

332

•4X

170

109

11

51

141

92

49

34

345

4X

169

110

10

49

152

87

65

36

358

4X

TITLE:

APPLICATimi:

REGION:

PLATFORM:

74

Final History and Forecast

PCB/Hybrid/MCM

Asia

Technical Workstation

©

t-*

P so

n

•o

c

I

•0

I g:

I

S

a a.

SP

•o ft

3

1986 1987

1988 1989 1990

CALCULATED AVERAGE SYSTEM PRICE DATA (Thousands of U.S.

Dollars)

Turnkey ASP 148.5 143.1

97.8 65.8

70.0

Hardware-Only ASP 24.5 24.3

14.5 11.8 13.2

1991

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 408 1,148

1,730

Workstation Shipments 408 1,148

1,730

CPU Installed Base 674 1.816

3,517

Workstation Installed Base 674 1,816

3,517

3,829

3,829

7,240

7,240

3,966

3,966

11,212

11,212

4,240

4,240

14,560

14,560

5,080

5,080

19,020

19,020

6,300

6,300

24,790

24,790

66.7

12.1

1992

64.0

11.1

1993

62.4

10.6

REVENUE DATA (Millions of U.S. Dollars)

Harckiare Revenue 28 79

81

CPU Revenue 21 57

53

Workstation Revenue 0 0

0

Peripheral Revenue (Turnkey) 7 22

29

Software Revenue 16 47

68

Bundled 14 39

51

Service Revenue 4 16

21

Total Revenue 49 143

170

Increase over Prior Year 8% 194%

20X

124

89

0

35

104

85

19

22

250

47%

133

98

0

35

109

81

28

24

267

7X

127

85

0

42

109

77

32

24

260

•2%

131

88

0

43

122

79

43

27

279

7%

135

92

0

43

136

78

59

30

301

8X

TITLE:

APPLICATION:

REGION:

PLATFORM:

75

Final History and Forecast

PCB/Hybrid/MCM

Asia

Host/Server

@

t - * o n

i

s

I

if

•o

SI

•3

I

I

n

I

5

3

cr

§:

1986 1987 1988 1989 1990 1991 1992 1993

UNIT SHIPMENT DATA ( W o r k » « t i o n ;Sihtj3ments)

CPU Shipflients 261 270 237 213 405 410 390 360

Workstation Shipments 953 835 816 705 843 800 730 640

CPU I n s t a l l e d Base 608 878 1,109 1,293 1,447 1,850 2,080 2,280

Workstation I n s t a l l e d Base 1,900 2,735 3,551 4,256 4,356 5,900 6,630 7,270

CALCULATED AVERAGE SYSTEM PRICE MTA (Thousands of U.S. D o l l a r s ) turnkey ASP 462.3 323.7 533.8 536.9 109.0 103.4 99.2 95.1

Hardware Only ASP 290.6 408.7 101.1 76.4 99.0 87.9 81.0 74.3

REVENUE DATA ( M i l l i o n s of U.S. D o l l a r s )

Hardware Revenue 80 72 67 35 32 30 28 24

CPU Revenue 32 33 33 21 22 13 12 10

Workstation Revenue 34 27 22 8 6 11 11 10

Peripheral Revenue (Turnkey) 14 12 12 5 4 5 5 4 software Revenue 29 22 30 14 12 11 9 7

Bundled 27 21 28 13 11 10 8 6

Unbundled 1 2 2 1 1 1 1 1

Service Revenue 11 11 15 8 6 6 6 5

Total Revenue 119 105 111 58 51 47 43 36

Increase over P r f W Year 12X -12X 6% -48X -12% -6% -9% -15%

TITLE:

APPLICATION:

REGION:

PLATFORM:

76

Final History and Forecast

PCB/Hybrid/MCM

Asia

Personal Co«nputer

•0

3

K

CT

I

©

t — •

!8

D

K

S

•g

n

f

a o.

if

n

3

S"

n

•O

I

1986 1987 1988

1989

1990 1991 1992

1993

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 971 3,006

Workstation Shipments 971 3,006

CPU Installed Base 1,518 4,482

3,515

3,515

7,851

Workstation Installed Base 1,518 4,482

7,851

2,000

2,000

9,417

9,417

2,047

2,047

10,132

10,132

1,980

1,980

10,730

10,730

1,960

1,960

10,990

10,990

CALCULATED AVERAGE SYSTEM PRICE DATA

Turnkey ASP

Hardware-Only ASP

(Thousands of U.S. Dollars)

34.4 18.6 12.0

2.3 6.1 7.9

15.2

6.0

16.8

5.3

16.3

5.2

16.1

5.2

1,810

1,810

11,170

11,170

15.9

5.1

REVENUE DATA (MiUtons of U.S. Dollars)

Hardware Revenue 9 21 28 23 19 13 12 10

CPU Revenue 6 19 18 20 17 9 8 7

Workstation Revenue 0 0 0 0 0 0 0 0

Peripheral Revenue (Turnkey) 3 2 9 2 2 3 3 3

Software Revenue 7 8 6 6 8 10 10 9

Bundled 4 5 1 2 4 5 4 4

Unbundled 3 3 5 4 4 5 5 5

Service Revenue 1 1 1 2 2 2 2 2

Total Revenue 17 30 35 31 29 25 23 21

Increase over Prior Year 127X 82% 16X -12X -6X -15X -7X -10X

TITLE:

APPLICATION:

REGION:

PLATFORM:

77

Final History and Forecast

PCB/Hybrid/MCM

Rest of World

All Platforms

@

o

u

S i

Wl

5"

8

1

3

I

if

•o

S

7

ar

•o

1

3

1

O"

1

1986 1987

""*""*"

UNIT SHIPMENT DATA (UorkstliSion

Shipments)

CPU Shipments

225

Workstation Shipments

CPU Installed Base

255

Workstation Installed Base

569

1,050

376

410

817

1,455

1988

= = S =

293

315

975

1.749

1989

S = = S

1990

====

385

458

1,293

2,143

427

529

1,660

2.642

1991

====

1992

====

570

710

1,920

3,030

790

970

2,480

3.790

1993

====

1,100

1,300

3,360

4.870

CALCULATED AVERAGE SYSTEM PRICE

DATA (Thousands of U.S. Dollars)

Turnkey ASP

Hardware-Only ASP

94.2

46.5

33.1

18.2

74.0

15.8

80.2

17.9

78.1

18.1

79.8

19.1

82.0

18.0

79.9

15.9

REVENUE DATA (NilUons of U.S. Dollars)

Hardware Revenue 11 7

CPU Revenue

8 5

Workstation Revertue

2

Peripheral Revenue (Turnkey) 1

1

Software Revenue

3

2 2

Bundled

Unbundled

Service Revenue

Total Revenue

Increase over PHisr Year

1

1

1

15

166X

1

1

1

10

-34X

6

5

1

0

2

1

1

1

9

-7%

8

6

2

0

1

1

1

2

11

20X

9

6

2

1

2

1

1

2

12

14%

11

6

4

1

3

2

1

3

17

36%

15

8

5

2

4

2

2

3

23

33%

18

10

5

2

6

3

4

5

29

29%

TITLE:

APPLICATION:

REGION:

PLATFORM:

78

Final History and Forecast

PC8/Hybrid/MCM

Rest of World

Technical Workstation

®

O u

n

I

3

•O

I g-

I

S ff a

fP

•o

I

1986 1987 1988

1989

1990 1991

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Installed Base 126 161 216

325

Workstation Installed Base 126 161 216

325

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP

Harduare-Only ASP

70.5 47.3 49.9

43.2

26.2 28.1 19.6

14.5

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue 3 1 2

2

CPU Revenue 3 1 1

2

Workstation Revenue 0 0 0

0

Peripheral Revenue (Turnkey) 0 0 0

0

Softuare Revenue 1 1 1

0

Bundled 1 0 0

0

Unbundled 0 0 0

0

Service Revenue 1

Total Revenue 5 2 3

3

Increase over Prior Year 162X -SSX 42X

15X

127

127

501

501

48.8

14.4

2

2

0

0

1

0

0

1

4

7X

200

200

570

570

46.6

13.1

5

41%

1992

340

340

860

860

44.7

12.0

2

1

1

2

4

4

0

1

8

57X

1993

570

570

1.380

1,380

43.6

11.4

7

5

0

1

3

1

2

3

13

60X

TITLE:

APPLICATION:

REGION:

PLATFORM:

79

Final History and Forecast

PCB/Hybrid/MCM

Rest of World

Host/Server

©

?

B

•s

n

en

I

3 a

if

"3

n n

•O

I

c

I

3

I a

I

CPU Shipments

Workstation Shipments

CPU Installed Base

Workstation Installed Base

1986

====

UNIT SHIPHEHT DATA (Workstation Shipments)

33

63

268

749

1987

====

11

45

155

794

1988

====

16

38

59

832

1989

====

25

99

81

931

CALCULATED AVERAGE SYSTEM PRICE

DATA (Thousands of U.S. Dollars)

Turnkey ASP

Hardware-Only ASP

396.7

217.5

590.9

408.0

771.2

145.9

743.9

156.8

1990

====

27

129

102

1,084

1991

====

40

190

140

1,250

1992

====

60

240

180

1,490

1993

====

70

260

240

1,750

474.6

170.7

453.3

155.4

434.8

142.7

416.6

130.8

REVENUE DATA <HiUions of U.S. Dollars)

Herduare Revenue

7 5

CPU Revenue

Workstation Revenue

4

3

3

2

Peripheral Revenue (Turnkey) 1

0

Software Revenue

Bundled

0

0

0

0

Unbundled

Service Revenue

Total Revenue

Increase over Prior Year

0

1

9

152X

0

0

5

-39%

3

2

1

0

0

0

0

1

4

-2A

5

3

2

0

0

0

0

1

6

52%

5

3

2

0

0

0

0

1

6

9%

7

2

4

1

1

1

0

1

9

42%

8

3

5

1

1

1

0

2

11

25%

9

3

5

1

1

1

0

2

12

11%

TITLE:

APPLICATION:

REGION:

PLATFORM:

80

Final History and Forecast

PCB/Hybrid/HCM

Rest of World

Personal ConfXJter

«

f

•o c n

§•

3

IT

•O

I

I

I

a a if

•a

H

3

1986 1987 1988 1989

1990

UNIT SHIPMENT DATA (Workstation Shipments)

CPU Shipments 99 330 217 237

273

Workstation Shipments 99 330 217 237

273

1991 1992 1993

330

330

1,220

1,220

390

390

1,440

1,440

470

470

1,740

1,740

CALCULATED AVERAGE SYSTEM PRICE DATA

(Thousands of U.S. Dollars)

Turnkey ASP

Hardware-Only ASP

24.8 3.3 14.6 19.8

18.0

6.6 4.7 6.3 5.4 5.4

17.8

5.3

17.6

5.3

17.5

5.2

REVENUE DATA (Millions of U.S. Dollars)

Hardware Revenue

CPU Revenue

Workstation Revenue

Peripheral Revenue (Turnkey)

Softusre Revenue

1

1

0

0

Bundled

Unbundled

Service Revenue

1

0

1

0

Total Revenue

Increase over Prior Year

2

298%

0

2

46X

1

0

1

1

1

0

0

1

1

0

0

1

0

1

0

2

-3X

1

1

0

0

0

0

0

0

2

-27X

2

1

0

0

1

0

1

0

2

42X

2

2

0

0

1

0

1

0

3

14X

2

2

0

0

1

0

1

0

3

18X

1

0

1

0

2

2

0

0

4

22X

Dataquest

Dataquest Research and Sales Offices:

Dataquest Incorporated

1290 Ridder Park Drive

San Jose, California 95131-2398

Phone: (408) 437-8000

Telex; 171973

Fax: (408) 437-0292

Technology Products Group

Phone: (800) 624-3280

Dataquest Incorporated

Ledgeway/Dataquest

The Corporate Center

550 Cochituate Road

Framingham, MA 01701

Phone: (508) 370-5555

Fax: (508) 370-6262

Dataquest Incorporated

Invitationa! Computer Conferences Division

3151 Airway Avenue, C-2

Costa Mesa, California 92626

Phone: (714) 957-0171

Telex: 5101002189 ICCDQ

Fax: (714) 957-0903

Dataquest Australia

Suite 1, Century Plaza

80 Berry Street

North Sydney, NSW 2060

Australia

Phone: (02) 959 4544

Telex: 25468

Fax: (02) 929 0635

Dataquest GmbH

Kronstadter Strasse 9

8000 Munich 80

West Germany

Phone: Oil 49 89 93 09 09 0

Fax: 49 89 930 3277

Dataquest Europe Limited

Roussel House, Broadwater Park

Denham, Uxbridge, Middx UB9 5HP

England

Phone: 0895-835050

Telex: 266195

Fax: 0895 835260/1/2

Dataquest Europe SA

Tour Gallieni 2

36, avenue du General-de-Gaulle

93175 Bagnolet Cedex

France

Phone: (1) 48 97 31 00

Telex: 233 263

Fax: (1) 48 97 34 00

Dataquest Hong Kong

Rm. 401, Connaught Comm. Bldg.

185 Wanchai Rd.

Wanchai, Hong Kong

Phone: 8387336

Telex: 80587

Fax: 5722375

Dataquest Israel

59 Mishmar Ha'yarden Street

Tel Aviv, Israel 69865 or

P.O. Box 18198

Tel Aviv, Israel

Phone: 52 913937

Telex: 341118

Fax; 52 32865

Dataquest Japan Limited

Shinkawa Sanko Building

1-3-17 Shinkawa, Chuo-ku

Tokyo 104 Japan

Phone: (03) 5566-0411

Fax: (03) 5566-0425

Dataquest Korea

Daeheung Bldg. 1105

648-23 "Vfeoksam-dong

Kangnam-gu

Seoul, Korea 135

Phone: (02) 556-4166

Fax: (02) 552-2661

Dataquest Singapore

4012 Ang Mo Kio Industrial Park 1

Ave. 10, #03-10 to #03-12

Singapore 2056

Phone: 4597181

Telex: 38257

Fax: 4563129

Dataquest Taiwan

Room 801/8th Floor

Ever Spring Building

147, Sect. 2, Chien Kuo N. Rd.

Taipei, Taiwan R.O.C. 104

Phone: (02) 501-7960

Telex: 27459

Fax: (02) 505-4265

Dataquest West Germany

In der Schneithohl 17

6242 Kronbeig 2

West Germany

Phone: 06173/61685

Telex: 418089

Fax: 06173/67901

0011354

i

nVlarket Share Estimates ^ ^ ^ ^ ^ ^ ^ ^ ^ |

Electronic Design Automation Applications ^ ^ ^ ^ ^ |

DataQuest

.

Source^

Dataquesl

m

1

k

Library Copy

DO NOT REMOVE!

^AD/CAM/(

;AE ^ I b

>

liNfORMATlOM RESOURCE CENTER

DATAQUESF INCORPORATED

1290 Ridder Pi^irk Drive

San Jose, CA 95131-2398

^408) 437-8600

Market Share Estimates

Electronic Design Automation Applications

Dataqyest CAD/CAM/CAE

Source:

Dataquest

i i

Published by Dataquest Incorporated

The content of this report represents our interpretation and analysis of information generally available to the public or released by knowledgeable individuals in the sut^ect industry, but is not guaranteed as to accuracy or completeness. It does not contain material provided to us in confidence by our clients.

Printed in the United States of America. All rights reserved. No part of this publication may be reproduced, stored in retrieval systems, or transmitted, in any form or by any means—mechanical, electronic, photocopying, duplicating, microfilming, videotape, or otherwise—without the prior permission of the publisher.

© 1991 Dataquest Incorporated I

June 1991

Table of Contents

Market Share Estimates.

Index of Tables

ELECTRONIC DESIGN AUTOMATION

Page

1 Market Database Methodology,

Table 33 Asia 89

ELECTRONIC CAE Table 53 Asia I l 6

i

Table 70 Europe Technical Workstation 141

Table 78 ROW Technical Workstation 153

i

K y

Market Share Estimates'Electronic

Design Automation Applications

Market Share Estimates

The following is a description of the information reported in the Market Share Estimates book for each segment:

Source—All companies in database; overview of industry

• Mechanical Applications—All companies in database with mechanical revenue

• AEC and GK Applications—All companies in database with AEC or CIS revenue

• Electronic Design Automation Applications—

All companies in database with EDA revenue

• Europe—All European-based companies and all other companies with more than $1 million in Eviropean revenue

• Asia—All Asian-based companies and all other companies with more than $1 million in Asian revenue

• Personal CAD and Distribution Chaimels—All companies in database with PC revenue

Market Database Methodology

Dataquesfs CAD/CAM/CAE industry service is committed to consistently delivering to clients the most accurate data available on the CAD/

CAM/CAE industry, both in terms of market history and forecast Our annual delivery schedule is designed to meet that commitment, as follows:

• First quarter January 31)—^We will provide complete preliminary market share tables for the year just completed. Historical database is opened for changes for a six-month period.

• Second quarter (March 31): We provide complete preliminary forecast tables, including a new five-year forecast period.

• Third quarter (July 1): We send complete final updated market share tables based on additional data collection and analysis over the previous six months as well as company history tables. At this point, the market share database is frozen and will not be changed until the end of the year. For the next six months, supplementary market data will be based on this final market data.

• Fourth quarter (October 1): We provide complete final forecast tables, taking into consideration changes in the market during the previous six months.

Since January, when the prelitninary market share numbers were published, over 35 new companies have been added to the market share database. This brings the total number of companies recognized in the CAD/CAM/CAE market to 430. The market share numbers also have been rechecked and verified for all vendors worldwide. Year-end 1990 financial results released this past spring have also been checked and incorporated into the final numbers.

Additionally, a number of companies significantly improved or made changes to their reporting methods during this past update cycle. These changes have been incorporated into this update. Dataquest's policy is to continually update its market information for any year, based on any new data received, in order to arrive at the most accurate market representation possible.

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Maritet Share Estimates—ElectroiUc Design Automatioa Applications

1

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Worldwide

Millions of Dollars/Actual Units

Company

Mentor Graphics

Sun

Digital

Cadence

Hewlett-Packard

Valid

Compaq

Dazix

IBM

Racal-Redac

Zuken

Fuj i tsu

NEC

Computervision

Intergraph

Seiko lnstrifnents--NO OEM

Siemens

Uchida Yoko

Wacom

Zycad

Scientific Calc.

Sharp System Products--NO OEM

Viewlogic Systems

Synopsys

Calay

EEsof

Autodesk

CAD IX

Teradyne

VLSI Technology

Apple Computer

Xilinx

ABB Cade International

=======

420.5

315.5

230.3

211.2

210.4

158.5

116.3

99.8

93.6

82.3

78.1

67.2

58.2

56.2

53.0

36.3

32.9

32.7

31.1

30.4

28.6

27.8

27.5

26.2

26.1

26.0

19.0

18.9

18.8

18.5

17.5

16.5

15.2 s:==s=ss s==s==s

120.8

275.7

163.9

.0

156.2

34.3

116.3

18.1

72.4

4.1

34.9

45.0

39,2

20.9

27.6

15.6

23.2

31.1

6.2

24.9

3.9

14.5

.0

.0

8.3

.1

.0

7.6

.0

.7

15.4

.0

8.9

176.6

.0

3.9

173.1

15.2

90.6

.0

39.9

10.3

66.3

43.2

15.5

14.4

15.1

9.3

16.5

11.5

1.6

21.7

.0

16.8

13.3

23.1

23.6

15.2

22.7

19.0

9.4

15.5

15.2

.0

14.8

4.6

= S S S = S =

4,167

16,852

1,941

0

14,016

945

19,377

735

11,011

380

552

1,492

3,581

442

558

232

437

519

509

250

102

166

0

0

425

18

0

139

0

53

3,360

0

66

= = = = S S =

13.3%

10.0%

7.3%

6.7%

6.6%

5.0%

3.7%

3.1%

3.0%

2.6%

2.5%

2.1%

1.8%

1.8%

1.7%

1.1%

1.0%

1.0%

1.0%

1.0%

.9%

.9%

.9%

.8%

.8%

.8%

.6%

.6%

.6%

.6%

.6%

.5%

.5%

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

=======

8.1% s=sssss

15.3%

=======

4.2%

18.6%

11.0%

.0%

.3%

17.1%

2.0%

.0%

15.0% .0%

10.5%

2.3%

7.8%

1.2%

4.9%

.3%

1.3%

7.8%

.0%

3.5%

.9%

5.7%

14.2%

1.0%

19.6%

.7%

11.2%

.4%

2.4%

3.0%

2.6%

1.4%

1.9%

1.1%

1.6%

2.1%

.4%

1.7%

.3%

1.0%

.0%

.0%

.6%

.0%

.0%

.5%

.0%

.0%

1.0%

.0%

.6%

3.7%

1.3%

1.2%

1.3%

.8%

1.4%

1.0%

.1%

1.9%

.0%

1.5%

1.2%

2.0%

2.0%

1.3%

2.0%

1.6%

.8%

1.3%

1.3%

.0%

1.3%

.4%

.6%

1.5%

3.6%

.4%

.6%

.2%

.4%

.5%

.5%

.3%

.1%

.2%

.0%

.0%

.4%

.0%

.0%

.1%

.0%

.1%

3.4%

.0%

.1%

(Continued)

01991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM;

REGION:

UNITS:

Matfcet Share EsUmatea—EkctroQlc D e s i ^ Automation Applications

1

1990 Final Market Share Estimates

Electronic Design Automation

Alt. Platforms

Worldwide

Millions of Dollars/Actual Units

Conpany

=======

LSI Logic

Ikos Systems

LPKF

SoIbourne

Logic Automation

Toshiba—NO OEM

Data I/O

Silvar-Lisco

Aucotec

Hitachi

ACTEL

Altera

Oread

Genrad

Analogy

CAD AM

Microsim

Kloeckner-Moeller

Ziegler

Quickturn Systems

Everex Systems

Meta-Softuare

Comdisco Systems

Ascent Logic Corp

Test Systems Strategies

Tokyo Electron--NO OEM

Seattle Silicon

CAD Software

BETRONEX

Vantage Analysis Systems

Sony

AnaCAO

Sophia Systems

Total

Revenue

=======

Hardware Software

Revenue Revenue

=======

=======

14.6

13.5

12.3

12.2

11.8

11.0

10.8

10.8

10.5

10.4

10.0

10.0

9.2

1.5

13.5

8.0

12.1

.0

5.5

.0

.0

3.7

5.0

.0

.0

.0

11.0

.0

3.1

.0

11.8

4.4

10.8

7.0

5.8

4.4

9.0

8.5

9.2

8.3

8.1

8.0

7.5

6.8

6.7

6.5

6.2

6.2

1.6

.0

3.5

.0

4.1

.0

6.5

6.2

.0

5.5

7.3

3.9

7.1

2.4

6.7

.0

.0

6.1

Ukstns

Shipped

=======

37

61

500

906

0

97

0

0

167

320

0

0

0

86

0

500

0

107

0

50

1,889

0

6.2

6.0

.0

.0

5.6 0

0

6.0

5.9

5.5

5.4

.0

2.0

.0

4.8

5.2

2.9

3.8

4.1

4.0

4.0

.0

.0

.0

4.6

4.1

4.0

.0

0

21

0

0

0

0

230

3.9

3.5

3.5

.0

.9

3.9

2.3

0 .

50

- Market

Total

Revenue

Revenue

=======

.5%

.4%

=======

.1%

.9%

Ukstns

Revenue

=======

Shipped

=======

.9X

.OX

.OX

.IX

.4%

.4%

.4%

.3%

.3%

.3%

.3%

.3%

.3%

.3%

.3X

.3%

.3X

.3%

.2%

.2%

.2X

.2%

.2%

.2%

.2X

.5%

.8%

.0%

.4%

.0%

.0%

.2%

.3%

.0%

.OX

.OX

.IX

.OX

.2X

.OX

.3X

.OX

.4X

.4X

.OX

.OX

.3X

.OX

1.0X

.4X

.9X

.6X

.5X

.4X

.8X

.7X

.8X

.5X

.6X

.3X

.6X

.2X

.6X

.OX

.OX

.5X

.5X

.5X

.9X

.OX

.IX

.OX

.OX

.2X

.3X

.OX

.OX

.OX

.IX

.OX

.5X

.OX

.IX

.OX

.IX

1.9X

.OX

.OX

.2X

.2%

.2%

.2%

.2X

.n

.IX

.n

.1%

.1%

.OX

.OX

.IX

.OX

.OX

.OX

.OX

.2X

.OX

.IX

.4X

.4X

.2X

.3X

.4X

.4X

.3X

.OX

.3X

.2X

.OX

.OX

.OX

.OX

.ox

.ox

.ox

.2X

.ox

.IX

i

(Continued)

i

01991 Dataqucst IiKorpoiated June—Seproduoioa Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estbnates—Electroalc Design Antamatloa AppUcatioas

1

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Worldwide

Millions of Dollars/Actual Units

Company

Quad Design Technology

EPIC Design Technology

Motorola

Pacific Nunerics

Quantic Laboratories

Assigraph

Academi Systems

Scientific & Engineering SW

Century Research Center debis Systentiaus

ALS Design

Research Machines

ALDEC

Infinite Graphics

Visionics

Accel Technologies

Integrated Silicon Systems

Dell Computer

Shared Resources

Royal Digital Systems

Ontologic

SIMUCAD

CAD-UL

CAD/CAM Group

Omation

Massteck

Sehluriberger

National Semiconductor

Emerald Design Systems

Phase Three Logic

Omron

DECAD

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

1.5

1.5

1.5

1.5

1.4

1.3

2.3

2.2

2.1

1.9

1.9

1.8

1.7

1.7

1.6

1.6

3.1

2.6

2.6

2.6

2.5

2.5

2.4

2.3

3.0

2.9

2.9

2.9

2.8

2.7

2.7

2.7

.1%

,1%

,0%

.ox

,0X

.ox

.ox

.ox

.ox

,1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.IX

.1%

.1%

.1%

11

24

0

39

5

0

0

0

16

0

0

0

0

0

21

11

0

0

0

0

0

15

90

0

396

0

0

10

0

15

558

4

.0

2.6

.0

.0

.0

.0

.4

2.2

.1

.0

.0

.0

.0

.0

.0

.0

.5

.3

.7

.0

1.2

.2

.0

.0

.0

.0

.0

.0

.3

.0

1.4

.7

.1

1,7

1.9

1.7

1.7

1.7

1.6

1.6

,5

1.2

.6

1.3

.1

.8

2.9

2.7

2.9

2.9

2.7

2.2

2,1

2.7

1.0

1.5

2.6

.0

2.5

2,5

1.8

2,1

1,9

.0

.ox

.ox

.2X

.ox

.ox

.ox

.ox

.ox

.2X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

,1X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.OX

.ox

.ox

.ox

•OX

.ox

.ox

.ox

,ox

,1X

.ox

,4X

,ox

.ox

.ox

.ox

.ox

.6X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.2X

.2X

.3X

.3X

.2X

.2X

.2%

.2X

.IX

.IX

.2X

.ox

.2X

.2X

.2X

.2X

.2X

.ox

.OX

.IX

.2X

.IX

.IX

.IX

.IX

.IX

.OX

.1X

.IX

.IX

.OX

.IX

(Continued)

01991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

1

1990 Final Market Share Estimates

Electronic Design Automation

All.Platforms

Worldwide

Millions of Dollars/Actual Units

Coopany

- — - —

Caditron

Cooper & Chyan Technology

Tanner Research

Spectrum Software

Serbi

Aucos elektronische Gerate

Intrinsix rotring euroCAD

Cadisys

Aptos

I SKA

MacNeaI-SchwendIer

Intercad

DAT Standard Information system

Nuniber One Systems

Inca

ICL

Douglas Electronics

Vamp

Foresight Resources

CAD Language Systems

Capilano Confuting

NCR Microelectronics

Object Design

Objectivity

PLUS Logic

The CAD Group

Imagine That

American Small Business Comp.

BV Engineering

Andor

- -- Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.5

.5

.5

.5

.4

.4

.4

.4

.4

.3

1.3

1.2

1.1

1.1

1.0

1.0

1.0

1.0

1.0

.9

.9

.8

.8

.8

.8

.7

.7

.6

.6

.5

.5

.6

.0

.0

.0

.0

.3

1.0

.5

.0

.0

.4

.0

.4

.0

.1

.0

.4

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.1

.4

.5

.4

.5

.5

.4

,4

.8

.2

.7

.7

.7

.2

.6

.6

.5

1.2

1.0

1.1

1.0

.7

.0

.4

1.0

.9

.4

.5 ss===ss

20

0

0

0

0

100

10

23

0

0

16

0

38

0

19

0

19

0

0

0

0

0

0

0

0

0

0

.4

.4

.4

.2

0

0

0

3 ss==sss:

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.ox

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

=ss==ss

.ox

sss==ss

.ox

=======

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.IX

.IX

.IX

.IX

.IX

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.IX

.ox

.IX

.ox

.IX

.IX

.IX

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

i

(Continued)

01991 Dataquest Incoiporated June—Repioductiai Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

1 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Worldwide

Millions of Dollars/Actual Units

Company

Microtel Pacific Research

Kontron Instruments

The Great Softwestern Co.

Bobcat Systems

Instrumatic Espanola

Simutest

Cascade Graphics

Innovative Data Design

DAPCO

Masta Corporation

Olivetti

Sierra Semiconductor

Kubota Computer

Dolphin Integration

Robocom

Other Companies

All Companies

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

=ss===s

.3

=======

.0

ss===s=

.2

=======

1

.2 .1 .1

4

.2 .0 .2

0

.2 .0 .2

0

.2 .0 .2

0

.2 .0 .2

0

.2

.2

.1

.1

.1

.1

.0

.0

.0

.0

.1

.0

.2

.2

.1

.1

.0

.1

0

0

0

1

17

0

• 1

• 1

.0

.1

.0

.0

.0

.1

.0

10

0

0

107.2

3,169.2

103.1 .3

1,484.7 1,155.9

9,849

98,690

.0%

.0%

.0%

.OX

.OX

.OX

.OX

.0%

.0%

.0%

.0%

.0%

.0%

.OX

.0%

3.4X

100.OX

.OX

.OX

.OX

.0%

.OX

.0%

.OX

.OX

.OX

.OX

.OX

.OX

.ox

.OX

.ox

6.9X

100.OX

.0%

.OX

.0%

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

100.OX

.OX

.OX

.OX

.OX

.OX

.OX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

10.ox

100.ox

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SU Companies

2,553.0

389.5

226.6

1,013.6

1,204.3

213.5

66.8

879.5

864.5

151.0

140.4

.0

87,853

7,976

2,862

78,240

80.6X

12.3X

7.2X

32. OX

81. IX

14.4X

4.5X

59.2X

74.8X

13.IX

12.IX

.OX

89.0X

8.1X

2.9X

79.3X

2,155.6 605.1 1,155.9 20,451

68.0X 40.8X 100.OX 20.7X

Source: Dataquest

June 1991

01991 Dataquest Iix:oiporated June—Reproducdon Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic D e s i ^ Antomation Applications

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

Worldwide

Millions of Dollars/Actual Units

I

Company

Mentor Graphics

Sun

Cadence

Hewlett-Packard

Valid

Dazix

Racal-Redac

Zuken

Digital

Conputervision

Intergraph

NEC

Siemens

Uchida Yoko

Seiko Instruments--NO OEM

Fujitsu

Calay

Zycad

Sharp System Products--NO OEM

Scientific Calc.

CADIX

VLSI Technology

ABB Cade International

LSI Logic

Synopsys

Ikos Systems

EEsof

Viewlogic Systems

Logic Automation

Toshiba—NO OEM

Silvar-Lisco

Teradyne

Analogy

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

420.5

276.7

185.9

173.9

158.5

99.8

74.3

74.2

62.0

55.8

39.7

35.3

32.9

32.7

32.4

27.6

26.1

24.3

24.2

23.4

18.9

17.1

15.2

13.1

12.6

12.6

11.7

11.6

11.6

11.0

10.3

7.9

7.7

15,2

6.4

15.2

.0

11.6

12.7

9.4

14.2

4.6

9.9

11.3

.0

9.8

9.8

11.6

4.4

6.6

6.6

6.9

176.6

.0

152.1

15.2

90.6

39.9

58.7

41.1

2.9

15.0

7.2

12.7

11.5

1.6

4,167

15,387

0

9,214

945

735

380

516

1,941

435

439

1,853

437

519

216

649

425

204

150

76

139

49

66

34

0

57

18

0

0

97

0

0

0

19.7

23.2

31.1

13.9

18.5

8.3

19.2

12.6

2.8

7.6

.5

8.9

1.3

.0

12.6

.1

.0

.0

5.5

.0

.0

.0

120.8

243.4

.0

126.8

34.3

18.1

4.1

33.0

41.9

20.8

21.0

1.1%

.9%

.8%

.7%

.6%

.6%

.6%

.5%

.5%

.5%

.5%

.5%

.4%

.4%

1.8%

1.6%

1.5%

1.5%

1.5%

1.3%

1.2%

1.1%

1.1%

19.3%

12.7%

8.5%

8.0%

7.3%

4.6%

3.4%

3.4%

2.8%

2.6%

20,0%

,0%

17.3%

1.7%

10.3%

4.5%

6.7%

4,7%

.3%

1,7%

,8%

1.4%

1.3%

.2%

1.7%

.7%

1.7%

.0%

1,3%

1,4%

1,1%

1,6%

.5%

1,1%

1.3%

.0%

1.1%

1.1%

1.3%

.5%

.8%

.7%

.8%

1.1%

1.1%

4.6%

1.1%

1,3%

.5%

1.6%

,1%

,0%

.0%

.0%

.2%

.0%

.0%

.0%

1.1%

.5%

.4%

.2%

.3%

.1%

,2%

.1%

,0%

10.3%

38.0%

.0%

22.8%

2.3%

1.8%

,9%

1,3%

4.8%

2.4%

2.2%

2.6%

3,5%

1.6%

2.1%

.9%

2.2%

1.4%

.3%

.9%

.1%

1.0%

.1%

.0%

1.4%

.0%

.0%

.0%

.6%

.0%

.0%

,0%

13.7%

27.6%

.0%

14.4%

3.9%

2.1%

.5%

3.7%

4.8%

2.4%

i

(Continued)

i

®1991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

T I T L E :

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Esdmates—Electronic O e s ^ Autonaatloa Applications

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

Worldwide

Millions of Dollars/Actual Units

Company

=======

Genrad

Quickturn Systems

Comdisco Systems

Ascent Logic Corp

Tokyo Electron--NO OEM

Seattle Silicon

Test Systems Strategies

Solbourne

Neta-Software

Hitachi

Vantage Analysis Systems

Sony

AnaCAD

Aucotec

Quad Design Technology

EPIC Design Technology

Motorola

Quantic Laboratories

IBM

Pacific Numerics

Scientific & Engineering SW

Century Research Center

Wacom

Shared Resources debis Systemhaus

Royal Digital Systems

Ontologic

Integrated Silicon Systems

Autodesk

CAD/CAM Group

Data I/O

ACTEL

SIMUCAD

Total

Revenue

=======

Hardware Software

Revenue

=======

Revenue

=======

Wkstns

Shipped

=======

5.2

5.2

4,0

4.0

3.9

3.5

3.1

3.0

7.2

6.5

6.2

6.0

5.9

5.5

5.4

5.4

2.9

2.9

2.8

2.8

2.7

2.7

2.5

2.1

2.0

1.9

1.9

1.7

1.7

1.7

1.6

1.5

1.5

1.4

6.5

.0

.0

2.0

.0

.0

5.3

.0

2,5

.0

3.5

.0

1.2

,0

.0

,0

,0

1.6

.0

.0

1.4

.6

.1

.5

.0

.0

.2

.0

.0

.0

.0

.0

4.8

.0

5.6

4.8

2.9

3.8

4.6

.0

5.2

2.2

4,0

,0

3,9

1.9

2.9

2.7

2.9

2.7

,6

2,8

2.7

1.0

1.9

,1

1.2

1.7

1.9

1.5

1.7

1.7

1.6

1.4

1.4

64

50

0

0

21

0

0

397

0

108

0

230

0

55

0

0

15

25

4

0

0

0

0

70

64

0

0

7

0

0

0

0

0

Total

Revenue

• narnei snare -•

Wkstns

'lardware Software

Revenue Revenue Shipped

=======

.2%

.2%

.2%

,2%

,2%

.2%

.3%

.3%

.3%

.3%

.3%

.3%

.2%

.2%

.OX

.3X

.OX

.4X

.OX

.IX

.OX

,2X

.7X

,0%

.OX

.2X

.OX

.OX

.6X

.5X

.OX

.6X

.5X

.3X

.4X

.5X

.OX

.6X

.2X

.5X

.OX

.4X

.2X

.3X

.2X

.IX

.OX

.OX

.IX

.OX

.OX

1.0X

.OX

.3X

.OX

.6X

.OX

.IX

.OX

,n

,1%

.1%

.1%

.1%

.1%

.n

.IX

.IX

.IX

.IX

,1X

.IX

.IX

.IX

.IX

.IX

.IX

.IX

.OX

.OX

.OX

.2X

.OX

.OX

.2X

.IX

.OX

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.2X

.2X

.2X

.2X

.2X

.2X

.2X

.2X

.3X

.3X

.3X

.IX

.3X

.3X

.IX

.2X

.OX

.IX

.OX

.OX

.OX

.2X

.OX

.OX

.OX

,1X

,ox

,2X

• OX

.OX

.OX

.ox

.OX

.ox

.ox

.ox

(Continued)

01991 Dataquest Incoiporated June—Reproduction Prohibited

10

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Maricet Share Estimalca—Electromic D e s t ^ Automatlcm AppUcadoas

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

Worldwide

Millions of Dollars/Actual Units

i

Conpany

Emerald Design Systems

Assigraph

Omron

National Semiconductor

Xilinx

DECAD

Sophia Systems

Cooper & Chyan Technology

Academi Systems

Schtinberger

I SKA

Microsim

ICL

Infinite Graphics

CAD Language Systems

Everex Systems

Object Design

NCR Microelectronics

Objectivity

Phase Three Logic

Inca

MacNeal-Schwendler

DAPCO

Sierra Semiconductor

Microtel Pacific Research

Serbi

Kubota Conputer

Dolphin Integration

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.8

.7

.5

.5

.5

.5

.5

.5

.4

.2

.2

.1

.1

.1

.1

.0

.0

1.5

1.4

1.4

1.3

1.3

1.3

1.3

1.2

1.1

1.1

.9

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.5

.0

.0

.0

.7

.0

1.2

.3

.0

.2

.3

.0

.1

.3

.4

.0

.4

.6

1.1

.1

1.1

1.2

.8

.8

1.2

.9

.4

.4

.8

.2

.5

.4

.0

.5

.4

.5

.4

.2

.2

.1

.1

.0

.1

.0

.0

24

11

39

11

0

4

10

0

4

9

16

0

19

0

0

60

0

0

0

0

0

0

0

0

0

0

7

0

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.0%

.OX

.OX

.OX

.ox

.ox

.0%

.ox

.ox

.ox

.ox

.ox

.ox

.0%

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.0%

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.IX

.ox

.IX

.IX

.IX

.IX

.IX

.IX

.ox

.ox

.IX

.ox

.IX

.ox

.ox

.IX

.ox

.IX

.ox

.ox

.0%

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

i

®1991 Dataquest Incorporated June—Reproduction Prohibited

(Continued)

i

>

I

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electron^ D e s ^ Automation AppUcatimis

2 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Teqhnical Workstation

Worldwide

Millions of Dollars/Actual Units

Company

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.0

2,177.1

.0

881.3

.0

881.6

0

40,474

.0%

100.0X

.0%

100.0%

.0%

100.0%

.0%

100.0%

1,734.5

279.0

163.5

464.5

680.7

153.3

47.2

397.8

669.7

111.4

100.5

.0

34,399

4,595

1,480

25,573

1,712.6 483.4 881.6 14,901

79.7X

12.8X

7.5%

21.3«

77.2%

17.4%

5.4%

45.1%

76.0%

12.6%

11.4%

.0%

85.0%

11.4%

3.7%

63.2%

Source: Dataquest

June 1991

11

>

01991 Dataquest IrKorporated June—fteproduction Probibited

1 2

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

1990 Final Market Share Estimates

Electronic Design Automation

Host/Server

Worldwide

Millions of Dollars/Actual Units

i

Company

Digital

IBM

Sun

Fujitsu

Cadence

Synopsys

Intergraph

Hewlett-Packard

Teradyne

Solbourne

Zycad

Scientific Gale.

Zuken

Seiko Instruments--NO OEM

Sharp System Products--NO OEM

Hitachi

LSI Logic

VLSI Techrwlogy

Assigraph

EEsof

Meta-Software

Genrad

MacNeaI-SchwendIer

Test Systems Strategies

Data I/O

Silvar-Lisco

CADAM

Schlimiserger

Analogy

Computervision debis Systentiaus

Logic Automation

Total

Revenue

Hardware Software

Revenue

Revenue

Wkstns

Shipped narnei anare -•

Total

Hardware Software

Revenue Revenue Revenue

Ukstns

Shipped

122.0

26.6

1.7

1.9

.8

.1

.2

.0

.0

.0

.2

.0

.0

.0

.0

32.2

19.4

.0

.0

6.6

6.4

.0

6.8

5.7

1.1

1.9

.0

.2

,0

.1

.1

.0

168.0

a.2

38.8

1.0

.8

.7

.6

.6

.5

,5

.4

.4

.4

.3

.2

28.9

25.3

«.6

13.3

7.7

7.5

6.8

6.1

5.2

3.9

3.9

3.6

1.7

1.5

1.4

1.4

1.3

.7

9.8

.0

6.7

20.9

12.3

2.1

.0

6.1

.0

.0

4.1

2.1

1.3

1.7

.7

1.1

1.0

1.1

1.2

.9

.6

.6

.6

.6

.4

.4

.1

.4

.1

.2

.2

0

543

1,464

489

0

0

119

388

0

509

46

27

35

16

16

138

3

3

11

0

0

9

0

0

0

0

0

7

0

8

6

0

35.5%

9.3%

8.2%

6.1%

5.4%

2.9%

2.8%

1.6%

1.6%

1.4%

.1%

.1%

.1%

.1%

.1%

.1%

.0%

.3%

,3%

.2%

.2%

.1%

.1%

.1%

1.3%

1.1%

.8%

.8%

.8%

.4%

.3%

.3%

.9%

12.5%

.0%

8.5%

26,7%

15.7%

2.7%

.0%

7,8%

.0%

.0%

5.2%

2.6%

1.7%

2.2%

.9%

1.4%

1.3%

1,4%

1,5%

1,2%

.7%

.8%

.8%

.7%

.5%

.5%

.2%

.5%

,1%

.2%

.3%

.0%

7.7%

20.7%

6.9%

.0%

.0%

1.7%

2.0%

.0%

.0%

.1%

.0%

.0%

,1%

.0%

.0%

.0%

.0%

.0%

.1%

.0%

.1%

.1%

.0%

5.5%

.0%

7.2%

.6%

.4%

.5%

.2%

.2%

39.1%

8.5%

10.3%

6.2%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.1%

.0%

.0%

.0%

2.1%

2.1%

.0%

2.2%

1.8%

.4%

.6%

.5%

.6%

.3%

.0%

.1%

i

(Continued)

i

©1991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

3 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Host/Server

Worldwide

Millions of Dollars/Actual Units

Company

Microsim

SIMUCAO

DECAO

Kubota Computer

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SU Companies

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.2

.2

.0

.0

82.1

473.1

= = s s s s

.0

.0

.0

.0

78.0

311.8

=======

.2

.2

.0

.0

.0

78.2

= = s s s s

0

0

1

3

3,242

7,083

.0%

.0%

.0%

.0%

17.4%

100.0%

.0%

.0%

.0%

.0%

25.0%

100.0%

.3%

.3%

.0%

.0%

.0%

100.0%

.0%

.0%

.0%

.0%

45.8%

100.0%

429.3

42.1

1.7

302.6

170.5

286.0

25.6

.1

245.4

66.4

64.4

12.5

1.3

.0

78.2

6,368

697

18

5,604

1,479

90.7X

8.9%

.4%

64.0%

36.0%

91.7%

8.2%

.0%

78.7%

21.3%

82.4%

16.0%

1.6%

.0%

100.0%

89.9%

9.8%

.3%

79.1%

20.9%

Source: Dataquest

Jme 1991

13

®1991 Dataquest Incoiporated June—Reproduction Piohibited

14

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Antomation Applications

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

Worldwide

Millions of Dollars/Actual Units

Company

Compaq

IBM

Hewlett-Packard

Wacom

NEC

Apple Computer

Autodesk

Viewlogic Systems

Xilinx

EEsof

LPKF

Fujitsu

Altera

Oread

Data I/O

ACTEL

Racal-Redac

CADAM

Aucotec

ICloeckner-Moel ler

Ziegler

Microsim

Everex Systems

CAD Software

BETRONEX

Hitachi

Teradyne

ALS Design

Research Machines

ALDEC

Visionics

Accel Technologies

Dell Computer

i

Total

Revenue

Hardware Software

Revenue Revenue

Wkstns

Shipped

Market Share

Total

Hardware Software

Revenue Revenue Revenue

Wkstns

Shipped

6.8

6.7

6.5

5.7

5.4

4.1

3.5

3.4

2.6

2.6

2.5

2.4

2.3

2.2

12.3

10.8

10.0

9.2

8.7

8.5

8.0

7.6

7.0

116.3

46.5

28.8

28.6

22.9

17.5

17.4

15.8

15.2

13.0

116.3

44.2

23.0

5.6

19,4

15.4

.0

.0

.0

.0

8.0

7.2

.0

.0

.0

.0

.0

3.5

2.5

4.1

.0

.0

5.7

.0

.0

1.7

.0

.0

2.6

.0

.0

.0

2.2

.0

.0

.0

19.9

1.7

.0

17.4

13.3

13.7

11.7

3.1

2.5

8.5

9.2

8.7

7.7

7.6

3.5

3.9

2.4

6,7

6.1

.0

4.6

4.1

1.5

2.8

2.6

.0

2.5

1.8

2.1

.0

19,377

10,399

4,414

484

1,728

3,360

0

0

0

0

500

354

0

0

0

0

0

500

112

107

0

0

1,828

0

0

74

0

0

396

0

10

0

558

1.4%

.0%

.0%

2.0%

.0%

.0%

.6%

.0%

.0%

.9%

.0%

.0%

.0%

.8%

2.5%

.0%

.0%

.0%

.0%

.0%

1.2%

.8%

39.9%

15.1%

7.9%

1.9%

6.7%

5.3%

.0%

.0%

.0%

.0%

2.7%

2.1%

1.9%

1.8%

1.7%

1.6%

1.5%

1.5%

1.4%

1.3%

1.3%

1.2%

1.1%

22.4%

9,0%

5.5%

5,5%

4,4%

3.4%

3.3%

3.1%

2,9%

2.5%

2.4%

1.0%

.8%

.7%

.7%

.5%

.5%

.5%

.5%

.4%

.4%

2.0%

1.2%

3.4%

3.1%

.0%

2.4%

2.1%

.8%

1.4%

1.3%

,0%

1.3%

.9%

1.1%

.0%

.0%

.0%

.0%

10.1%

.9%

.0%

8.9%

6.8%

7,0%

6.0%

1,6%

1,3%

4,3%

4.7%

4.4%

3.9%

3.9%

1.8%

37.9%

20.3%

8.6%

.9%

1.0%

.2%

.2%

.0%

.0%

3.6%

.0%

.0%

.1%

.0%

.0%

.8%

.0%

.0%

.0%

1.1%

3.4%

6.6%

.0%

.0%

.0%

.0%

1.0%

.7%

,0%

,0%

.0%

.0%

.0%

i

(Continued)

i

01991 Dataquest IrKorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic D e s ^ Automation Applications

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

Worldwide

Millions of Dollars/Actual Units

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Company

Sophia Systems

Infinite Graphics

CAD-UL

Omation

Academi Systems

Massteck

Caditron

Tanner Research

Spectrum Software

Phase Three Logic

Aucos etektronische Gerate

Intrinsix

Serbi rotring euroCAD

Cadisys

Ikos Systems

Apt OS

Intercad

DAT Standard Information system

Nuiiber One Systems

Douglas Electronics

Integrated Silicon Systems

Vamp

Foresight Resources

Capilano Computing

Inca

PLUS Logic

Digital

The CAD Group

Imagine That

American Small Business Comp.

BV Engineering

=======

2.2

=======

.6 sssss=s

1.5

=======

40

2.0

1.7

.0

.0

2.0

1.7

0

0

1.6

1.6

1.6

1.3

1.1

1.1

1.1

1.0

1.0

1.0

1.0

1.0

1.0

.9

.8

.8

.8

.6

.6

.6

.5

.5

.5

.4

.4

.4

.4

.4

.4

.0

.2

.0

.6

.0

.0

.0

.3

1.0

.0

.5

.0

1.0

.0

.4

.0

.1

.0

.2

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

1.6

1.3

1.6

.5

1.0

1.1

1.0

.7

.0

1.0

.4

1,0

.0

.9

.2

.7

.7

.6

.4

.6

.5

.5

.5

.4

.3

.4

.4

.4

.4

0

7

0

20

0

0

0

100

10

0

23

0

4

0

38

0

19

0

8

0

0

0

0

0

0

0

0

0

0

=======

.4%

=======

.2X

=======

.7X

=======

.IX

.4%

.3%

.OX

.OX

1.0X

.9X

.OX

.3%

.3%

.3%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.IX

.1%

.1%

.IX

.IX

.IX

.OX

.IX

.OX

.2X

.OX

.OX

.OX

.IX

.3X

.OX

.2X

.OX

.3X

.OX

.IX

.OX

.OX

.OX

.IX

.OX

.OX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.8X

.7%

.8X

.3X

.5X

.6X

.5X

.4X

.OX

.5X

.2X

.5X

.2X

.3X

.2X

.3X

.2X

.2X

.OX

.5X

.IX

.4X

.4X

.3X

.2X

.2X

.2X

.2X

.2X

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.2X

15

(Continued)

®1991 Oataquest Incorporated June—Reproduction Prohibited

16

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

4 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Personal Coirputer

UorldHide

Millions of Dollars/Actual Units

i

Company

Andor debis Systemhaus

Genrad

Kontron Instrunents

The Great Softwestern Co.

Bobcat Systems

Instrumatic Espanola

Microtel Pacific Research

Simutest

Cascade Graphics

Innovative Data Design

National Semiconductor

Pacific Numerics

Masta Corporation

Olivetti

SIMUCAO

Meta-Software

Dolphin Integration

Robocom

NCR Microelectronics

Other Companies

All Companies

All U.S.-Based Conpanies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

.1

.1

.1

.0

.2

.2

.2

.1

.0

.0

. 0

25.1

519.0

.3

.3

.2

.2

.2

.2

.2

.2

. 2

.0

.0

.0

.0

.0

25.1

291.7

.0

.0

.0

.0

.0

.0

.1

.0

.0

.0

.0

.1

.1

.0

.1

17

0

0

0

0

0

6,606

51,133

0

0

0

0

0

1

0

0

0

1

3

20

12

4

.1

.0

.1

,0

.0

.0

.0

.3

196.1

.2

.2

. 2

.2

.2

.2

.2

.1

.2

.2

.2

.2

.1

389.2

68.4

61.4

246.6

272.4

237.6

34.6

19.5

236.4

55.3

130.4

27.2

38.6

.0

196.1

47,087

2,683

1,363

47,062

4,071

.0%

.0%

.0%

.0%

.0%

8.6%

100.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

. 1 %

. 1 %

.OX

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

4.8%

100.0%

. 1 %

. 1 %

. 1 %

.0%

.0%

.0%

.0%

.0%

.0%

. 1 %

. 1 %

. 1 %

. 1 %

. 1 %

. 1 %

. 1 %

. 1 %

. 1 %

. 1 %

.0%

.2%

100.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

12.9%

100.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

75.0%

13.2%

11.8%

47.5%

52.5%

81.5%

11.9%

6.7%

81.0%

19.0%

66.5%

13.9%

19.7%

.0%

100.0%

92.1%

5.2%

2.7%

92.0%

8.0%

Source: Dataquest

June 1991

i i

®1991 Dataquest Incorporated June—Reproduction Prohibited

Company

Mentor Graphics

Sun

Cadence

Digital

Hewlett-Packard

Valid

Compaq

Dazix

Intergraph

IBM

Racal-Redac

Viewlogic Systems

Zycad

Computervision

EEsof

Synopsys

Scientific Calc.

Xilinx

Apple Computer

Teradyne

Autodesk

Calay

SoIbourne

VLSI Technology

LSI Logic

ACTEL

Ikos Systems

Logic Automation

Ascent Logic Corp

Altera

Everex Systems

Microsim

M.eta-Software

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Esdixutes—Electronic D c s ^ Automation Apjdicatlons

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

North America

Millions of Dollars/Actual Units

Market Share --

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

12.3

10.7

10.7

10.5

10.1

9.1

9.0

8.9

8.8

8.2

8.1

7.7

6.0

5.8

5.6

4.9

4.7

31.9

26.7

21.9

20.1

19.8

16.7

15.1

14.4

203.2

164.8

116.9

113.2

85.7

«4.6

54.7

38.5

58.9

144.0

.0

78.9

64.5

12.1

54.7

7.7

17.4

20.9

1.2

.0

16.2

4.6

.1

.0

1.8

.0

9.4

.0

.0

2.9

9.0

.4

.9

.0

8.1

.0

.0

.0

5.6

.0

.0

85.3

.0

86.8

3.9

6.7

56.4

.0

15.4

5.6

2.8

17.8

16.9

.0

4.1

13.2

13.0

7.3

9.7

.0

8.7

10.1

5.3

.0

7.3

6.6

7.4

.0

7.7

4.8

4.9

.0

4.6

4.6

2,013

9,118

0

932

5,555

494

9,107

284

336

3,281

98

0

163

110

10

0

44

0

2,050

0

0

149

669

25

22

0

37

0

0

0

1,700

0

0

10.0%

24.6%

.0%

13.5%

11.0%

2.1%

9.3%

1.3%

3.0%

3.6%

.2%

.0%

2.8%

.8%

.0%

.0%

.3%

.0%

1.6%

.0%

.0%

.5%

1.5%

.1%

.2%

.0%

1.4%

.0%

.0%

.0%

1.0%

.0%

.0%

.7%

.7%

.7%

.6%

.6%

.6%

.5%

.4%

.4%

.4%

.4%

2.4%

2.0%

1.6%

1.5%

1.5%

1.3%

1.1%

1.1%

15.3%

12.4%

8.8%

8.5%

6.4%

6.4%

4.1%

2.9%

.9%

.8%

.8%

.8%

.8%

.7%

17.1%

.0%

17.4%

.8%

1.3%

11.3%

.0%

3.1%

1.1%

.6%

3.6%

3.4%

.0%

.8%

2.6%

2.6%

1.5%

1.9%

.0%

1.7%

2.0%

1.1%

.0%

1.5%

1.3%

1.5%

.0%

1.5%

1.0%

1.0%

.0%

.9%

.9%

4.8%

21.8%

.0%

2.2%

13.3%

.0%

.0%

.4%

1.6%

.1%

.1%

.0%

.1%

.0%

.0%

.0%

4.1%

.0%

.0%

1.2%

21.8%

.7%

.8%

7.8%

.2%

.0%

.4%

.3%

.0%

.0%

.1%

.0%

4.9%

17

(Continued)

01991 Dataquest Incorporated June—Reproduction Prohibited

18

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Mailcet Share Estimates—Electronic Design Autmnation Applications

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

North America

Millions of Dollars/Actual Units

Company

Quickturn Systems

Data I/O

Comdisco Systems

Zuken

Genrad

Oread

Analogy

Test Systems Strategies

Seattle Silicon

Quad Design Technology

Silvar-Lisco

Vantage Analysis Systems

CAD Software

EPIC Design Technology

Pacific Numerics

Infinite Graphics

Motorola

Visionics

CAD AM

Quantic Laboratories

Shared Resources

Dell Computer

Scientific & Engineering SU

ALDEC

Accel Technologies

LPKF

Ontologic

CAD/CAM Group

Royal Digital Systems

SIMUCAD

Emerald Design Systems

Phase Three Logic

Integrated Silicon Systems

.0

.0

.0

.0

.0

.0

.0

1.1

.0

4.6

.0

.0

1.7

.7

.0

.0

.0

.0

.0

.0

.1

1.9

.0

.0

.0

1.2

.0

.0

.0

.0

.7

.0

.3

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

2.1

1.9

1.9

1.9

1.8

1.8

1.8

1.7

2.8

2.7

2.7

2.6

2.5

2.5

2.4

2.4

2.3

4.6

4.3

4.2

3.9

3.7

.3.7

3.6

3.6

3.3

3.1

2.8

1.6

1.5

1.5

1.5

1.4

.2%

.2%

.1%

.1%

.1%

.U

.1%

.1%

.1X

.IX

.IX

.IX

.IX

.IX

.3%

.3%

.35;

.3%

.3%

.3%

.3%

.3%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

2.1

.1

.0

1.9

1.9

1.7

.5

1.8

2.8

2.3

2.4

2.6

2.5

2.5

1.8

1.2

.0

4.3

3.7

2.1

2.5

3.7

3.3

3.1

2.3

2.9

1.7

1.7

1.5

1.4

.6

1.3

1.1

0

0

10

150

0

4

474

0

0

0

0

0

0

0

0

0

35

0

0

27

39

0

0

0

0

75

0

0

0

0

24

0

9

.IX

.OX

.ox

.1X

.IX

.OX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.4X

.ox

.ox

1.1X

.ox

.ox

.ox

.2X

.ox

.ox

.ox

.ox

.IX

.ox

.0%

.ox

.9X

.8X

.4X

.5X

.7X

.7X

.6X

.5X

.6X

.3X

.6X

.5X

.5X

.5X

.5X

.5X

.4X

.2X

.4X

.ox

.ox

.4X

.4X

.3X

.IX

.4X

.3X

.3X

.3X

.IX

.3X

.2X

.ox

.ox

.ox

.2X

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.2X

.ox

.ox

.3X

.ox

.ox

.8X

.OX

.OX

.3X

.IX

.OX

.OX

.OX

.OX

.OX

.ox

.OX

(Continued)

01991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic D e s ^ Automation Applications

1990 Final Market Share Estimates

Electronic Design Automation

AIt.Platforms

North America

Millions of Dollars/Actual Units

Company

Cooper & Chyan Technology

BETRONEX

Massteck

Omation

Intrinsix

Spectrin) Software

Cadisys

National Semiconductor

Tanner Research

Apt OS

Schlumberger

Douglas Electronics

Vamp

Capilano Computing

PLUS Logic

NCR Microelectronics

Academi Systems

Object Design

Foresight Resources

Objectivity

The CAD Group

BV Engineering

American Small Business Comp.

Imagine That

CAD Language Systems

The Great Softwestern Co.

Bobcat Systems

Simutest

Hicrotel Pacific Research

Innovative Data Design

Cascade Graphics

1.2

1.1

1.1

1.1

1.0

1.0

1.0

.9

.9

.7

.7

.6

.6

.5

.4

.4

.4

.4

.4

.4

.4

.4

.3

.3

.3

.2

.2

.2

.2

.2

.2

.0

.0

.0

.0

1.0

.0

.0

.2

.0

.0

.2

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

2

0

0

0

0

0

0

0

0

0

0

0

10

0

0

7

0

0

0

0

0

0

7

0

1

0

0

0

0

0

0

.4

.3

.4

.3

.4

.4

.3

.6

.6

.5

.4

1.2

1.1

1.1

1.1

.0

1.0

1.0

.7

.8

.7

.4

.3

.3

.3

.2

.2

.2

.2

.2

.2

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.IX

.IX

.IX

.IX

.IX

.IX

.IX

.1X

.IX

.1%

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.2X

.2X

.2X

.2X

.ox

.2X

.2X

.IX

.2X

.IX

.IX

.IX

.IX

1%

1%

1%

IX

1%

ox

,0%

,0%

.0%

,0%

,0%

,0%

1%

1%

,1%

,1%

1%

,1%

,0%

,0%

,0%

,0%

,0%

.0%

.0%

.ox ox ox ox ox ox

.ox

.ox

.ox

.ox

.2X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

»

(Continued)

©1991 Dataquest Incorporated June—Reproduction Prohibited

20

TABLE NUMBER:

TITLE:

APPLICATION:

PUTFORM:

REGION:

UNITS:

Market Share Estimates—Electronic D e s ^ Automation Applications

5 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

A l l Platforms

North America

Millions of Dollars/Actual Units

i

Company

Sierra Semiconductor

Other Companies

All Companies

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.1

55.9

1,328.5

.0

53.7

586.5

.1

.3

497.6

0

4,735

41,806

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SU Companies

1,290.2

3.9

34.4

491.8

579.4

1.7

5.3

428.5

470.5

2.1

25.0

.0

41,455

27

323

36,904

836.7 157.9 497.6 4.902

.0%

4.2%

100.0%

.0%

9.2%

100.0%

.0%

. 1 %

100.0%

.0%

11.3%

100.0%

97.1%

.3%

2.6%

37.0%

98.8%

.3%

.9%

73.1%

94.6%

.4%

5.0%

.0%

99.2%

. 1 %

.8%

88.3%

63.0% 26.9% 100.0% 11.7%

Source: Dataquest

June 1991

i

®1991 Dataquest Incotpotated June—Reproduction Prohibited

i

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Etectronlc Design Antomation Applications

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

North America

Millions of Dollars/Actual Units

Company

Mentor Graphics

Sun

Cadence

Valid

Hewlett-Packard

Dazix

Digital

Intergraph

Racal-Redac

Coirputervision

Zycad

Scientific Calc.

Calay

Viewlogic Systems

LSI Logic

VLSI Technology

Ikos Systems

Logic Automation

Synopsys

EEsof

Ascent Logic Corp

SoIbourne

Quickturn Systems

Teradyne

Comdisco Systems

Meta-Software

Zuken

Analogy

Seattle Silicon

Genrad ,

Test Systems Strategies

Quad Design Technology

Silvar-Lisco

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

58.9

127.2

.0

12.1

53.8

7.7

20.4

13.4

1.2

4.5

12.5

1.3

2.9

.0

.8

.2

7.5

.0

.0

.1

.0

4.6

4.6

.0

.0

.0

1.6

.0

.0

.6

.0

.0

.0

203.2

144.5

103.5

84.6

72.4

38.5

31.7

23.9

19.1

16.6

15.8

9.9

9.1

6.8

6.0

4.6

4.6

4.4

4.2

8.5

7.9

7.8

7.5

7.5

6.9

3.9

3.7

3.5

3.3

3.3

3.2

3.1

2.8

85.3

.0

76.9

56.4

6.7

15.4

2.9

4,3

15.1

4.1

.0

5.3

5.3

7.1

5.9

6.5

.0

7.5

6.2

5.7

4.8

.0

.0

3.7

3.7

3.9

2.0

3.1

2.3

2.2

2.8

2.9

1.7

2,013

8,324

0

494

3,847

284

932

265

98

109

133

32

149

0

20

23

34

0

0

29

0

0

0

0

0

10

0

338

35

0

0

0

26

17.4%

37.6%

.0%

3.6%

15.9%

2.3%

6.0%

4.0%

.3%

1.3%

3.7%

.4%

.9%

.0%

.2%

.1%

2.2%

.0%

.0%

.0%

.0%

1.3%

1.3%

.0%

.0%

.0%

.5%

.0%

.0%

.2%

.0%

.0%

.0%

1.1%

1.0%

.9%

.9%

.9%

.8%

.8%

.8%

.7%

.7%

.5%

.5%

.5%

.5%

.4%

.4%

.4%

.4%

.4%

.4%

.3%

.3%

22.3%

15.8X

11.3X

9.3X

7.9%

4.2%

3.5%

2.6%

2.1X

1.8%

1.7%

22.4%

.0%

20.2%

14.8%

1.8%

4.1%

.8%

1.1%

4.0%

1.1%

.0%

1.4%

1.4%

1.9%

1.6%

1.7%

.0%

2.0%

1.6%

1.5%

1.3%

.0%

.0%

1.0%

1.0%

1.0%

.5%

.8%

.6%

.6%

.7%

.7%

.4%

11.6%

48.1%

.0%

2.9%

22.2%

1.6%

5.4%

1.5%

.6%

.6%

.8%

.2%

.9%

.0%

.1%

.1%

.2%

.0%

.0%

.1%

.0%

2.0X

.2%

.0%

.0%

.0%

.1%

.0%

.0%

.2%

.0%

.OX

.0%

21

(Continued)

01991 Dataquest Incorporated June—fieproduction Prohibited.

22

TABLE NUMBER:

T I T L E :

APPLICATION:

PLATFORM:

REGION:

UNITS:

Mariiet Share Estimates—Electronic Design Antomadon AppUcatioas

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

North America

Millions of Dollars/Actual Units

Company

Vantage Analysis Systems

EPIC Design Technology

Pacific Nunerics

Motorola

Quantic Laboratories

Shared Resources

Scientific & Engineering SW

Ontologic

CAD/CAM Group

Royal Digital Systems

Emerald Design Systems

SIMUCAD

ACTEL

Cooper & Chyan Technology

Integrated Silicon Systems

Autodesk

Xilinx

National Semiconductor

IBM

Data I/O

Microsim

Infinite Graphics

Schlunberger

Everex Systems

NCR Microelectronics

Phase Three Logic

Object Design

Objectivity

CAD Language Systems

Academi Systems

Sierra Semiconductor

Microtel Pacific Research

- Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

2.8

2.7

2.5

2.5

2.3

.9

.9

.8

.8

.7

.5

.5

.5

.5

.2.1

1.9

1.8

1.7

1.6

1.5

1.3

1.2

1.2

1.0

.4

.4

.4

.4

.3

.2

.1

.0

.2

.0

.4

.4

.4

.4

.3

.1

.1

.0

1.7

1.5

.6

1.2

1.1

1.2

.9

2.8

2.4

2.5

2.5

2.1

.1

1.9

1.8

.9

.8

.6

.2

.7

.5

.5

.0

.0

.2

.5

.0

.0

.0

.2

.5

.0

.0

.0

.7

.0

.0

.0

.1

.0

.0

.0

.0

.0

.1

.0

.0

.0.:

.0

-.0

.0

.0

.0

.0

.0%

.0%

.2%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.1%

.0%

.0%

.0%

.1%

.1%

.0%

.0%

.0%

.ox

.ox

.ox

.ox

.ox

.OX

.2X

.IX

.1%

.OX

.OX

.IX

.4X

.2X

.3X

.3X

.3X

.2X

.2X

.2X

.2X

.IX

.IX

.1%

.IX

.OX

.OX

.OX

.7X

.6%

.7X

.6X

.6X

.OX

.5X

.5X

.4%

.1%

.1%

.1%

.OX

.OX

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.2%

.2%

.2%

.1%

.1%

.1%

.1%

.3%

.3%

.3%

.3X

.3%

.2%

.2%

.2%

.1%

.1%

.1%

.1%

.1%

4

54

0

0

0

0

0

1

0

0

0

0

24

0

0

0

4

0

0

4

0

0

0

0

0

0

0

7

19

0

0

0

.OX

.OX

.ox

.OX

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.0%

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.3X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

(Continued)

®1991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Autmnatlofi Applications

6 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

North America

Millions of Dollars/Actual Units

23

Coirpany

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

.0

912.2

.0

338.0

.0

380.1

0

17,312

.0%

100.0%

.0%

100.0%

.0%

100.0%

.0%

100.0%

880.1 332.2 357.6 17,039

3.7 1.6 2.0 26

.28.4 4.1 20.5 247

229.7 198.4 .0 12,788

682.5 139.7 380.1 4,525

96.5% 98.3% 94.1% 98.4%

.4% .5% .5% .1%

3.1% 1.2% 5.4% 1.4%

25.2% 58.7% .0% 73.9%

74.8% 41.3% 100.0% 26.1%

Source: Dataquest

June 1991

®1991 Dataquest Incotporated June—Reproduction Prohibited

24

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

1990 Final Market Share Estimates

Electronic Design Automation

Host/Server

North America

Millions of Dollars/Actual Units

i

Company

Digital

Sun

Cadence

IBM

Intergraph

Synopsys

SoIbourne

Teradyne

Zycad

Hewlett-Packard

Scientific Gale.

VLSI Technology

LSI Logic

EEsof

Meta-Software

Genrad

Test Systems Strategies

Data I/O

Schlumberger

Zuken

Analogy

SIMUCAD

Computervision

Logic Automation

Nicrosim

CAOAM

81,1

20.3

13.4

.4

.4

.2

.2

.2

.2

.2

.2

.2

.2

.1

11.9

8.0

7.5

4.5

4.2

4.0

3.3

2.5

1.1

.9

.8

.7

58.5

16.8

.0

7.2

4.0

.0

4.4

.0

3.7

2.7

.5

.1

.1

.0

.0

.1

.0

.0

.1

.1

.0

.0

.0

.0

.0

.0

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.7

.0

1.9

.8

.7

.7

.7

.3

.0

9.9

2.6

1.3

6.7

.0

3.4

.0

.4

.2

.1

.1

.2

.2

.0

.2

.1

.1

0

793

0

142

72

0

331

0

30

163

12

3

2

0

0

4

0

0

3

2

0

0

1

0

0

0

1.2%

.5%

.4%

.4%

.3%

.2%

.2%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

38.5%

9.6%

6.3%

5.7%

3.8%

3.6%

2.1%

2.0%

1.9%

1.5%

.0%

23,9%

.0%

4,3%

2.2%

.0%

10.0%

,0%

,9%

4,9%

,4%

.1%

.1%

.0%

.0%

.1%

.0%

.0%

.1%

.1%

,0%

,0%

.0%

.0%

.0%

.0%

2.2%

.0%

31.8%

8.4%

4.1%

21.6%

.0%

10,9%

.0%

.0%

6,2%

2,6%

2.1%

2.2%

2.1%

.8%

1,2%

.7%

,2%

.3%

.5%

,5%

,1%

,5%

,4%

.4%

41.6%

12.0%

.0%

5.1%

2.8%

.0%

3.1%

.0%

2.6%

1.9%

.4%

.1%

.1%

.0%

.0%

.0%

.0%

.0%

.0%

.1%

.0%

.0%

.0%

.0%

.0%

.0%

i

©1991 Dataquest Incorporated June—Reproductiaa Prohibited

(Continued)

i

TABLE NIMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Maiicet Share Estimates—Electronic Design Automation Applications

7 (Continued)

1990 Final Market Share Estimates

Electronic Design Autoination

Host/Server

North America

Millions of Dollars/Actual Units

Conpany

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SU Companies

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

21.2%

100.0X

30. U

100.0%

.0%

100.0%

53.1%

100.0%

44.6

210.9

210.7

. 2

.0

152.8

58.1

42.4

140.8

140.7

.1

.0

124.9

15.9

.0

31.2

1,762

3,320

31.1

.1

.0

.0

31.2

3,319

2

0

3,050

270

99.9X

. n

.ox

72.5X

27.5%

99.9%

. 1 %

.0%

88.7%

11.3%

99.7%

.3%

.0%

.0%

100.0%

99.9%

. 1 %

.0%

91.9%

8.1%

Source: Dataquest

June 1991

25

01991 Dataquesc Incoiporated June—Reproduction Prohibited

Company

Compaq

IBM

Viewlogic Systems

Apple Computer

Hewlett-Packard

Xilinx

Autodesk

EEsof

ACTEL

Altera

Everex Systems

Microsim

Oread

Data I/O

Racal-Redac

CAD Software

Visionics

CADAM

Infinite Graphics

Dell Computer

Teradyne

ALDEC

Accel Technologies

LPKF

BETRONEX

Massteck

Oroation

Phase Three Logic

Intrinsix

Spectrum Software

Cadisys

Tanner Research

Aptos

26

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electrcwdc Des^^n Autamatiaa ^plications

8

1990 Final Market Share Estimates

Electronic Design Automation

Pers.onal Computer

North America

Millions of Dollars/Actual Units

.0

.0

1.2

.0

.0

.0

.0

.0

.0

.0

.0

1.1

.0

1.9

.0

54.7

13.3

.0

9.4

8.1

.0

,0

.0

.0

.0

5.2

.0

.0

1.0

.0

.0

.0

.0

1.9

1.8

1.8

1.1

1.1

1.1

1.1

1.0

1.0

1.0

.9

.7

54.7

U.O

11.6

10.7

10.1

9.9

9.2

7.5

7.0

5.8

5.2

4.2

3.7

3.5

2.9

2.7

2.4

2.3

2.0

1.9

1.9

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

1.9

1.7

.5

1.1

1.1

1.1

1.0

.0

1.0

1.0

.8

.7

.0

.0

9.7

.0

.0

8.9

9.2

6.8

6.3

4.9

.0

3.9

3.7

3.5

2.7

2.3

1.8

1.0

2.0

.0

1.6

9,107

3,120

0

2,050

1,545

0

0

0

0

0

1,645

0

0

0

0

0

10

150

0

474

0

0

0

75

0

0

0

0

10

0

0

0

0

.0%

.0%

1.1%

.0%

.0%

.0%

.0%

.9%

.0%

.0%

.0%

.0%

50.8%

12,3%

.0%

8.7%

7.5%

.0%

.0%

.0%

.0%

.0%

4.8%

.0%

.0%

.0%

.0%

.0%

.0%

1.0%

.0%

1.8%

.0%

.9%

.9%

.5%

.5%

.5%

,5%

.5%

.5%

.5%

.4%

.3%

26.6%

6.8%

5.6%

5.2%

4.9%

4.8%

4.5%

3.7%

3.4%

2.8%

2.5%

2.0%

1.8%

1.7%

1.4%

1.3%

1.2%

1.1%

1.0%

.9%

.9%

.9%

2.2%

.0%

.0%

.0%

.4%

.0%

.0%

,0%

.0%

.0%

.0%

.0%

.0%

.0%

43.0%

14.7%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.7%

.0%

9.7%

7.3%

.0%

.0%

.0%

.0%

.0%

7.8%

2.2%

1.9%

.5%

1.3%

1,3%

1.2%

1.1%

.0%

1.2%

1.1%

.9%

.8%

.0%

.0%

11.3%

.0%

.0%

10.3%

10.7%

7.9%

7.3%

5.7%

.0%

4.6%

4.3%

4.0%

3.1%

2.7%

2.1%

1.2%

2.3%

.0%

1.8%

(Continued)

©1991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

8

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

North America

Millions of Dollars/Actual Units

Company

Douglas Electronics

Ikos Systems

Vamp

Capilano Computing

PLUS Logic

Digital

Foresight Resources

The CAD Group

BV Engineering

Integrated Silicon Systems

American Small Business Comp.

Imagine That

Academi Systems

The Great Softwestern Co.

Bobcat Systems

Simutest

Innovative Data Design

Cascade Graphics

Microtel Pacific Research

Pacific Numerics

Genrad

National Semiconductor

SIMUCAD

Meta-Software

NCR Microelectronics

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.6

.6

.6

.5

.4

. .4

.4

.4

.4

.4

.3

.3

.2

.2

.2

.2

,2

.2

.2

.1

.1

.1

.1

.0

.0

.4

.2

.3

.3

.2

.2

.2

.2

.2

.2

.6

.D

.6

.5

.4

.3

.3

.4

.0

.0

.0

.0

.0

.0

.0

.0

.1

.0

.0

.0

.0

.0

.0

.0

.0

.0

.6

.0

.0

.0

.0

.0

.0

2X

2%

2X

2X

1%

1%

3%

3X

3%

2%

2X

2%

2X

2%

1%

1%

1%

1%

1%

1%

n

0%

OX

OX

OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.2X

.2X

.2X

.2X

.2X

.IX

.IX

.IX

.OX

.OX

.7X

.ox

.6X

.5X

.5X

.4X

.4X

.4X

.4X

.3X

.4X

.4X

.2X

.3X

.3X

.0%

.5X

.0%

.OX

.OX

.OX

.OX

.OX

.OX

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

27

01991 Dataquest Incorporated June—Reproduction Prohibtted

(Continued)

28

TABLE NUMBER:

T I T L E :

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

8 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

North America

Millions of Dollars/Actual Units

Company

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

— Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

11.3

205.5

11.3

107.7

.3

86.3

2,973

21,173

5.5%

100.0%

10.5%

100.0%

.3%

100.0%

14.0%

100.0%

199.4

.0

. 6 . 1

109.3

96.2

106.5

.0

1.2

105.3

2.4

81.8

.0

4.5

.0

86.3

21,097

0

76

21,066

107

97.1%

.0%

2.9%

53.2%

46.8%

98.9%

.0%

1.1%

97.8%

2.2%

94.8%

.0%

5.2%

.0%

100.0%

99.6%

.0%

.4%

99.5%

.5%

Source: Dataquest

June 1991

i

(

®1991 Dataquest Incorporated June—Reproduction Prohibited

i

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Esllniiilrj ThLUiwifc Dcdga Antomatkm AppUcadoos

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Europe

Millions of Dollars/Actual Units

Company

Mentor Graphics

Hewlett-Packard

Sun

Digital

Compaq

Dazix

Racal-Redac

Valid

IBM

Computervision

Siemens

Cadence

Intergraph

ABB Cade International

Calay

Aucotec

LPKF

Scientific Calc.

EEsof

VLSI Technology

Teradyne

Kloeckner-Moeller

Ziegler

Autodesk

Apple Computer

Zycad

Synopsys

AnaCAD

Genrad

Oread

LSI Logic

Analogy

Viewlogic Systems

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

119.5

95.0

92.2

90.1

54.6

44.9

39.7

39.6

35.6

34.0

31.5

31.3

18.4

15.2

14.4

4.6

3.9

3.9

3.7

3.7

3.6

3.2

3.0

10.5

9.9

9.5

7.8

7.6

7.3

6.8

6.6

6.5

4.7

22.5

.0

8.9

8.9

4.5

3.7

6.4

1.2

.0

.2

.0

4.1

.0

.0

4.1

3.8

.0

.0

.7

.0

.4

.0

.0

33.5

69.4

80.5

65.3

54.6

7.2

1.6

2.7

27.8

14.0

11.0

23.3

3.2

4.6

8.3

5.8

2.5

5.6

6.8

50.2

6.1

.0

.0

.0

18.0

32.2

23.8

3.8

8.3

6.3

6.0

2.4

6.6

6.5

.0

.0

3.5

3.9

2.5

3.7

2.8

2.9

2.5

1,184

6,542

3,968

776

9,107

331

181

204

4,397

278

418

0

194

66

234

167

400

34

5

22

0

107

0

0

907

38

0

0

39

0

9

0

0

1.9%

1.9%

1.0%

.8%

1.4%

.2%

.0%

.0%

.0%

.9%

.0%

.0%

.9%

.8%

.0%

.0%

.2%

.0%

.1%

.0%

.0%

7.1%

14.7%

17.1%

13.9%

11.6%

1.5%

.3%

.6%

5.9%

3.0%

4.8%

.0%

1.1%

1.0%

1.0%

.8%

.8%

.8%

.7%

.7%

.7%

.5%

.5%

.4%

.4%

.4%

.4%

.4%

.3%

.3%

12.5%

9.9%

9.6%

9.4%

5.7%

4.7%

4.1%

4.1%

3.7%

3.5%

3.3%

3.3%

1.9%

1.6%

1.5%

16.3%

2.0%

.0%

.0%

.0%

5.8%

10.5%

7.8%

1.2%

2.7%

3.6%

7.6%

1.0%

1.5%

2.7%

1.9%

.8%

1.8%

2.2%

2.0%

2.0%

.8%

2.2%

2.1%

.0%

.0%

1.2%

1.3%

.8%

1.2%

.9%

.9%

.8%

1.2%

.0%

.6%

.2%

.7%

.5%

1.2%

.1%

.0%

.1%

.0%

.3%

.0%

.0%

2.7%

3.5%

19.4%

11.8%

2.3%

27.0%

1.0%

.5%

.6%

13.1%

.8%

.1%

.0%

.0%

.1%

.0%

.0%

.0%

.0%

29

(Continued)

®1991 Daiaquest Incoiporated June—KepnxfaiGtion Prohibited

30

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Europe

Millions of Dollars/Actual Units

Company

Xilinx

Logic Automation

Silvar-Lisco

Assigraph

Ikos Systems

Altera debis Systemhaus

Solbourne

ALS Design

Research Machines

Academi Systems

BETRONEX

CAD-UL

Microsim

Data I/O

ACTEL

CAD AM

DECAD

Caditron

CAD Software

Test Systems Strategies

Comdisco Systems

Serbi

Aucos elektronische Gerate

Vantage Analysis Systems rotring euroCAD

I SKA

MacNeal-Schwendler

Schlianberger

Intercad

DAT Standard Information system

Number One Systems

Inca

2.2

1.7

1.7

1.6

1.6

1.6

1.3

1.3

1.3

3.0

3.0

2.9

2.8

2.7

2.6

2.6

2.6

2.6

2.6

2.3

1.2

1.2

1.0

1.0

1.0

1.0

.9

.8

.8

.8

.8

.8

.7

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

1.1

1.0

1.1

1.0

.7

1.0

.4

.4

.8

.2

.2

.7

.7

.7

1.8

2.2

1.7

1.6

1.6

1.4

.8

.8

.5

2.7

3.0

1.8

2.2

.0

2.2

1.5

.0

2.6

.0

.0

.2

.4

.0

.1.

.0

.0

.0

.0

.0

.3

.0

.5

.4

.2

.0

.0

.0

.0

.0

.7

.2

.6

.0

.0

.0

.0

2.7

.0

.7

2.6

.0

2.6

0

0

0

21

12

0

90

192

0

396

9

0

0

0

0

0

100

5

20

0

0

0

0

100

0

23

16

0

8

38

0

19

0

.3%

.3%

.3%

.3%

.3%

.3%

.3%

.3%

.3%

.3%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.1%

.IX

.IX

.9X

1.0X

.6X

.7%

.ox

.7X

.5X

.ox

.8X

.ox

.6X

.7X

.6X

.5%

.5X

.5X

.3X

.3X

.2X

.3%

.3X

.3X

.3X'

.2X

.3X

.IX

.IX

.3X

.IX

.IX

.2X

.2X

.2X

.OX

.5X

.OX

.OX

.OX

.OX

.0%

.OX

.OX

.OX

.6X

.OX

.IX

.5X

.OX

.OX

.IX

.OX

.IX

.OX

.OX

.OX

.ox

.IX

.ox

.IX

.IX

.ox

.ox

.IX

.ox

.ox

.ox

1.2X

.OX

.OX

.OX

.OX

.OX

.OX

.3X

.OX

.OX

.OX

.IX

.OX

.OX

.3X

.6X

.OX

.OX

.IX

.OX

.OX

.OX

.ox

.3X

.OX

.IX

.ox

.ox

.ox

.IX

.ox

.IX

.ox

(Continued)

®1991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Maricet Share Estimates—Electronic Design Automation Appficadom

1990 Final Market Share Estimates

Electronic Design Automation

A11. Platforms

Europe

Millions of Doliars/Actual Units

Company

ICL

Quickturn Systems

National Semiconductor

Seattle Silicon

Scientific & Engineering SU

Omation

Zuken

ALDEC

Everex Systems

Massteck

Accel Technologies

Motorola

Dell Computer

Pacific Numerics

Seiko Instruments--NO OEM

Kontron Instrtinents

Instrimatic Espanola

Meta-Software

Tanner Research

Foresight Resources

Aptos

CAD Language Systems

DAPCO

Integrated Silicon Systems

Masta Corporation

Olivetti

Ontologic

Microtel Pacific Research

Imagine That

Objectivity

Dolphin Integration

.1

.0

.0

.0

.0

.0

.3

.0

.1

.1

.5

.0

.0

.0

.4

.7

.1

.0

.0

.0

.2

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

0

1

1

1

1

1

1

2

2

1

1

1

1

1

1

2

2

3

4

0

3

5

5

3

5

2

0

5

4

0

4

0

0

0

0

0

0

0

0

0

84

0

2

4

0

19

5

4

0

0

0

4

0

151

1

1

17

0

0

0

0

0

1%

1%

1%

IX

1%

1%

1%

1%

1%

0%

OX

OX

OX

ox ox ox ox ox ox ox ox ox ox ox ox ox ox ox ox ox ox

.ox

.IX

.ox

.ox

.IX

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

,2X

.IX

.2X

.2X

.IX

.2X

.ox

.IX

.IX

.IX

.IX

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.4X

.ox

.0%

.ox

.2X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

31

(Continued)

®1991 Dataquest Incorpotated June—Reproduction Prohibited

32

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

9 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Europe

Millions of Dollars/Actual Units

i

Company

Cascade Graphics

American Small Business Comp.

Object Design

NCR Microelectronics

Robocoffl

Other Companies

All Companies

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

.0

.0

.0

.0

.0

31.5

.0

.0

.0

.0

.0

30.3

.0

.0

.0

.0

.0

.0

0

0

0

0

0

2,728

.0%

.0%

.0%

.0%

.0%

3.3%

.0%

.0%

.0%

.ox

.0%

6.4%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

8.1%

959.3 471.0 307.4 33,677 100.0% 100.0% 100.0% 100.0%

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

792.4 412.3 210.9 31,320

82.6% 87.5% 68.6% 93.0%

.8 .4 .4 5

166.1 58.3 96.1 2,352

. 1 % . 1 % . 1 % .0%

17.3% 12.4% 31.3% 7.0%

364.7 313.8 .0 28,035

38.0% 66.6% .0% 83.2%

594.6 157.2 307.4 5,642

62.0% 33.4% 100.0% 16.8%

i

Source: Dataquest

June 1991

®1991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Antomatlon Applications

10

1990 Final Market Share Estimates

Electronic Design Autonfiation

Technical Workstation

Europe

Millions of Dollars/Actual Units

Company

Mentor Graphics

Sun

Hewlett-Packard

Dazix

Valid

Racal-Redac

Computervision

Siemens

Cadence

Digital

ABB Cade International

Calay

Intergraph

Scientific Calc.

VLSI Technology

AnaCAD

Zycad

EEsof

Aucotec

LSI Logic

Genrad

Analogy

Teradyne

Silvar-Lisco

Logic Automation

Ikos Systems debis Systemhaus

Synopsys

Assigraph

Viewlogic Systems

DECAD

Comdisco Systems

Test Systems Strategies

Total

Revenue

Hardware Software

Revenue Revenue

Ukstns

Shipped

Total 1

• napKei snare -• hardware Software

Ukstns

Revenue Revenue Revenue Shipped

.2

.0

2.9

.0

1.2

.3

.6

.0

.0

.0

.0

2.5

.5

.0

.0

.0

.2

.0

.0

33.5

71.1

54.1

7.2

2.7

1.6

13.9

22.5

.0

16.5

8.9

4.5

6.6

.8

3.3

3.3

3.1

3.1

2.9

2.9

2.5

2.0

1.9

1.4

1.3

1.3

1.2

1.1

119.5

80.8

76.0

44.9

39.6

34.5

33.8

31.5

27.5

23.3

15.2

14.4

13.8

7.8

7.5

3.9

3.7

3.5

3.5

50.2

.0

6.1

18.0

23.8

27.3

8.2

11.0

20.4

.0

4.6

8.3

2.5

4.3

1.8

2.9

.0

1.2

1.7

1.1

1.1

.8

1.1

.9

6.2

3.9

.0

3.0

1.9

2.5

2.2

2.8

2.6

1,184

3,617

3,951

331

204

181

271

418

0

776

66

234

153

25

22

0

31

5

55

8

29

0

0

0

0

11

64

0

11

0

4

0

0

1.1%

.0%

.5%

.1%

.2%

.0%

.0%

.0%

.0%

1.0%

.2%

.0%

.0%

.0%

.1%

.0%

.0%

13.1%

27.8%

21.2%

2.8%

1.0%

.6%

5.4%

8.8%

.0%

6.5%

3.5%

1.8%

2.6%

.3%

.1%

.0%

.5%

.4%

.3%

.3%

.2%

.2%

.2%

.2%

.2%

1.2%

.6%

.6%

.6%

.6%

.5%

.5%

.5%

.5%

.5%

19.0%

12.9% -

12.1%

7.2%

6.3%

5.5%

5.4%

5.0%

4.4%

3.7%

2.4%

2.3%

2.2%

1.2%

21.9%

.0%

2.7%

7.8%

10.4%

11.9%

3.6%

4.8%

8.9%

.0%

2.0%

3.6%

1.1%

1.9%

2.7%

1.7%

.0%

1.3%

.8%

1.1%

.9%

1.2%

1.1%

.8%

1.3%

.0%

.5%

.7%

.5%

.5%

.3%

.5%

.4%

.2%

.0%

.0%

.0%

.0%

.1%

.5%

.0%

.1%

1.3%

.2%

.2%

.0%

.3%

.0%

.5%

.1%

.0%

.0%

.0%

.0%

10.1%

30.7%

33.6%

2.8%

1.7%

1.5%

2.3%

3.6%

.0%

6.6%

.6%

2.0%

33

(Continued)

©1991 Daoquest Incorporated June—Reproduction Prohibited

34

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design AutiMnation Applications

10

1990 Final Market Share Estimates

Electronic Design Automation

Tecirnical Workstation

Europe

Millions of Dollars/Actual Units

i

Company

IBM

Vantage Analysis Systems

Academi Systems

I SKA

ICL

Quickturn Systems

Autodesk

Schlunberger

Seattle Silicon

Scientific & Engineering SU

National Semiconductor

SoIbourne

Zuken

Motorola

Pacific Numerics

Seiko Instruments--NO OEM

ACTEL

Data I/O

Xilinx

Inca

Nicrosim

MacNeal-Schwendler

Meta-Software

CAD Language Systems

DAPCO

Integrated Silicon Systems

Ontologic

Objectivity

Serbi

Everex Systems

Dolphin Integration

Object Design

•Q

P.1

.9

.0

*ii

.5

.2

.0

.0

.,t

^ti

Jt

,0

,6

.5

.0

.1

.4

.4

.7

.&

.6

.0

J>

.0

•>!!>

.Q

.d

JB

4

4

. Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.2

.1

.1

.1

.1

.1

.0

.0

.0

.2

.2

.2

.2

.2

.2

.2

1.0

1.0

.9

.9

.7

.7

.6

.6

.6

.5

.5

.5

.5

.4

.3

.3

0

0

0

0

2

0

0

0

0

0

4

37

3

0

0

27

0

4

16

19

5

0

5

0

0

0

0

0

0

5

0

.2

.2

.2

.2

.1

.3

.1

.2

.2

.2

.2

1.0

.7

.4

.2

.0

.6

.2

.4

.5

.4

.0

.3

.4

.1

.1

.1

• 1

.1

.0

.0

.0

Q

2%

2%

1%

IX

1%

1%

n n

IX

IX

IX

IX

IX

IX

ox

OX

ox ox ox ox ox ox ox ox ox ox ox ox ox ox ox ox

.IX

.4X

.3X

.2X

.IX

.ox

.2X

.IX

.2X

.2X

.2X

.ox

.IX

.2X

.IX

.IX

.IX

.IX

.IX

.IX

.IX

.1X

.IX

.IX

.ox

.ox

.0%

.ox

.ox

.ox

.ox

.ox

.2X

.ox

.ox

.IX

.2X

.ox

.ox

.ox

.ox

.ox

.ox

.3X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.2X

.ox

.ox

.2X

.2X

.3X

.ox

.1X

.ox

.ox

.ox

.2X

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

i

(Continued)

i

01991 Oataquest Incorporated June—Reproduction Prohibited

\

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estitnirtrs Flectronic D e s ^ Automatton AppUcattons

10 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

Europe

Millions of Dollars/Actual Units

>

Company

Microtel Pacific Research

NCR Microelectronics

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SU Companies

Market Share

Total Hardware Software Ukstns Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped Revenue Revenue Revenue Shipped

.0

.0

.0

627.6

.0

.0

.0

255.5

.0

.0

.0

229.7

0

0

0

11,777

.0%

.0%

.0%

100.0X

.0%

.0%

.0%

100.0%

.0%

.0%

.0%

100.0%

.0%

.0%

.0%

100.0%

516.4

.7

110.5

154.1

473.5

214.9

.3

40.3

129.7

125.8

167.6

.4

61.7

.0

229.7

10,701

5

1,071

7,689

4,089

82.3%

.IX

17.6%

24.6X

75.4%

84.1%

.1%

15.8%

50.8%

49.2%

73.0%

.2%

26.9%

.0%

100.0%

90.9%

.0%

9.1%

65.3%

34.7%

Source: Dataquest

June 1991

35

^

01991 Dataquest Incorpotated June—Reproduction Prohibited

36

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Marlcet Share Estbnalies—ElectronJc D e s ^ Automati0a Applications

11

1990 Final Market Share Estimates

Electronic Design Automation

Host/Server

Europe

Millions of Dollars/Actual Units

Company

=======

Digital

IBM

Sun

Intergraph

Cadence

Hewlett-Packard

Teradyne

Solbourne

Synopsys

Scientific Calc.

Assigraph

Zycad

MacNeaI-SchwendIer

EEsof

Genrad

LSI Logic debis Systemhaus

Schluiterger

Computervision

Analogy

Test Systems Strategies

CAD AM

Data I/O

VLSI Technology

Logic Automation

Microsim

DECAO

Meta-Software

Seiko Instruments--NO OEM

Zuken

?1

.1

.0

.0

.0

.0

-Z

>2

^2

.1

-1

.1

,1

66.8

16.0

11.4

4.6

3.8

3.2

2.9

2.1

2.0

1.7

1.4

.9

.7

.4

4

.4

.3

48.8

9.6

9.4

2.3

.0

2.7

.0

2.1

.0

.3

.0

.9

.0

.0

.1

.0

.1

.1

.1

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.0

3.5

.0

.7

2.8

.0

2.4

.0

1.8

1.3

1.1

.0

.6

.4

.3

.3

.2

.1

.0

.2

.1

.1

.1

.0

.1

.1

.0

.0

.0

.0

0

211

351

41

0

163

0

155

0

9

11

7

0

0

^4

1

0

0

0

0

6

3

7

0

0

0

1

0

0

0

1.0%

.6%

.4%

.3%

.3%

.2%

.2%

.2%

.1%

46.1%

11.0%

7.8%

3.2%

2.6%

2.2%

2.0%

1.4%

1.4%

1.2%

.1%

.1%

.1%

.1%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

2.2%

1.5%

1.7%

1.1%

.4%

.2%

.9%

.7%

.5%

.5%

.2%

.4%

.3%

.1%

.2%

.1%

.1%

.0%

21.9%

.0%

4.5%

17.5%

.0%

14.8%

.0%

11.4%

8.1%

6.8%

.0%

3.9%

48.7%

9.6%

9.4%

2.3%

.0%

2.7%

.0%

2.1%

.0%

.3%

.0%

.9%

.0%

.0%

.1%

.0%

.1%

.1%

.1%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.OX

.0%

.0%

.0%

.0%

10.8%

18.0%

2.1%

.0%

8.3%

.0%

7.9%

.0%

.4%

.5%

.3%

.0%

.0%

.2%

.0%

.3%

.2%

.3%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

(Continued)

®1991 Dataquest Incorporated June—Reproduction Prohibited

>

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estltnates—Electronic Design Automation Applications

11 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Host/Server

Europe

Millions of Dollars/Actual Units

37

Company

Other Companies

All Companies

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

24.9

145.0

23,7

100.1

.0

16.1

984

1,955

17.2%

100.0%

23.6%

100.0%

.0%

100.0%

50.3%

100.0%

All U.S.-Based Companies 143.3 100.0 14.8 1,937 98.8% 99.9% 91.9% 99.1%

All Asian-Based Companies .1 .0 .0 0 .0% .0% .1% .0%

All European-Based Companies -1.7 .1 1.3 18

1.2% .1% 8.0% .9%

All Hardware Companies 108.4 86.7 .0 1*,653

74.8% 86.5% .0% 84.6%

All Turnkey & SU Companies 36.6 13.5 16.1 301

25.2% 13.5% 100.0% 15.4%

Source: Dataquest

June 1991

^

I

01991 Dataquest Incorpotated June—Reproduction Prohibited

38

TABLE NUMBER:

T I T L E :

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Elcctmolc Deslga Antomatioa Applications

12

1990 Final Market Share Estimates

Electronic Design Automation

Personal Conputer

Europe

Millions of Dollars/Actual Units

i

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Company

Compaq

IBM

Hewlett-Packard

LPKF

Aucotec

Kloeckner-Moeller

Ziegler

Autodesk

Racal-Redac

Apple Computer

EEsof

Oread

Xilinx

Altera

ALS Design

Research Machines

BETRONEX

Viewlogic Systems

CAD-UL

CADAM

Microsim

Academi Systems

ACTEL

Teradyne

Data I/O

Caditron

CAD Software

Aucos elektronische Gerate

Serbi rotring euroCAD

Intercad

DAT Standard Information system

Nunnber One Systems

=======

54.6 sssssss

54.6

18.6

15.8

9.9

7.0

.6.8

6.6

5.9

5.2

4.7

3.9

3.7

2.7

2.6

2.6

2.6

2.2

1.8

1.7

1.5

1.4

1.4

1.4

1.3

1.3

1.3

1.3

1.0

1.0

1.0

.8

.8

.8

17.7

12.6

6.4

2.5

4.1

.0

.0

.0

4.1

.0

.0

.0

.0

.0

2.6

.0

.0

.0

.7

.0

.1

.0

.0

.0

.6

.0

.3

.0

.5

.4

.0

.1

=======

.0 sssssss

9,107

.0

.0

2.5

3.9

2.4

6.6

5.9

4.9

.0

4,159

2,428

400

112

107

0

0

0

907

3.5

3.7

2.5

2.2

2.6

.0

2.2

1.5

1.7

.7

1.3

1.1

1.2

1.1

1.3

.5

1.1

.7

1.0

.4

.2

.7

.7

0

0

0

0

0

396

0

0

0

100

0

6

0

0

0

20

0

100

0

23

38

0

19

29.3%

10.0%

8.5%

5.3%

3.8%

3.6%

3.6%

3.2%

2.8%

2.5%

2.1%

2.0%

1.5%

1.4%

1.4%

1.4%

1.2%

.9%

.9%

.8%

.8%

.7%

.7%

.7%

.7%

.7%

.7%

.5%

.5%

.5%

.4%

.4%

.4%

.6%

.0%

.1%

.0%

.0%

.0%

.6%

.0%

.3%

.0%

.0%

.0%

.0%

2.2%

.0%

.0%

.0%

47.4%

15.3%

11.0%

5.6%

2.1%

3.5%

.0%

.0%

.0%

3.6%

.0%

.0%

.5%

.3%

.0%

.0%

.0%

.0%

.0%

4.0%

6.3%

3.8%

10.8%

9.6%

8.0%

.0%

5.7%

6.0%

4.0%

3.6%

4.2%

.0%

3.5%

2.4%

2.8%

1.1%

2.2%

1.8%

2.0%

1.8%

2.1%

.8%

1.7%

1.1%

1.6%

.7%

.4%

1.2%

1.1%

4.5%

.0%

.0%

.0%

.0%

.0%

2.0%

.0%

.0%

.0%

.5%

.0%

.0%

.0%

.0%

.0%

.1%

.0%

.5%

.0%

.1%

.2%

.0%

.1%

45.7%

20.9%

12.2%

2.0%

.6%

.5%

.0%

.0%

.0%

i

(Continued)

01991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automatioti Applications

12

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

Europe

Millions of Dollars/Actual Units

Company

=======

Omation

ALDEC

Inca

Everex Systems

Massteck

Accel Technologies

Dell Computer debis Systemhaus

Kontron Instruments

Instrunatic Espanola

Ikos Systems

Tanner Research

Foresight Resources

Aptos

Genrad

Masta Corporation

Olivetti

Imagine That

Microtel Pacific Research

National Semiconductor

Cascade Graphics

American Small Business Comp.

Integrated Silicon Systems

Dolphin Integration

Robocom

Pacific Nunerics

Total Hardware Software Wkstns

Revenue

=======

Revenue

=======

Revenue

=======

.5

.5

.5

.5

.4

.4

.3

.3

.2

.2

.2

.2

.2

.2

.1

.1

p1

*1

.0

.0

.0

.5

.0

.0

.3

.1

.1

.0

.2

.0

.0

.0

.0

.0

.1

.0

.5

.5

.5

.0

.4

.3

.0

.2

.1

.2

.0

.1

.2

.2

.1

.1

.0

.1

Shipped

=======

0

0

0

146

0

0

84

20

4

0

1

0

0

0

6

1

17

0

• 1

.1

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.1

.1

.0

.0

.0

.0

.0

.0

0

0

0

0

0

0

0

0

- Market Share -

Total

Hardware Software

Revenue

=======

Revenue

=======

Revenue

=======

.3%

.OX

.9X

Wkstns

Shipped

=======

.OX

.3%

.OX

.8X

.OX

.3%

.2X

.2%

.2%

.2%

.2%

.n

.1%

.n

.1%

.1%

.n

.IX

.1%

.1%

.ox

.4X

.ox

.OX

.3X

.IX

.IX

.ox

.2X

.OX

.ox

.OX

.ox

.ox

.IX

.8X

.OX

.6X

.5X

.OX

.3X

.2X

.3X

.OX

.2X

.2X

.2X

.IX

.IX

.OX

.IX

.ox

.7X

.OX

.OX

.4X

.IX

.OX

.OX

.OX

.OX

.ox

.ox

.ox

.ox

.IX

.0%

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.IX

.IX

.OX

.OX

.OX

.OX

.OX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

39

01991 Dataquest Incoiporated June—Reproduction Prohibited

(Continued)

40

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Maitet Share Estimates—Electronic Design Automatioa Applications

12 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Personal Co(nputer

Europe

Millions of Dollars/Actual Units

Company

Other Conpanies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

6.6

186.7

6.6

115.3

.0

61.6

1,744

19,945

3.6%

100.0%

5.7%

100.0%

.0%

100.0%

8.7%

100.0%

132.8

.0

53.9

102.2

84.5

97.4

.0

17.9

97.5

17.9

28.5

.0

33.1

.0

61.6

18,682

0

1.262

18,693

1,252

71.1%

.0%

28.9%

54.7%

45.3%

84.5%

.0%

15.5%

84.5%

15.5%

46.2%

.0%

53.8%

.0%

100.0%

93.7%

.0%

6.3%

93.7%

6.3%

Source: Dataquest

June 1991

i i

01991 Dataquest Incoiporated June—Sepioduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Maricet Share Estimates—tSteuoaic Design Automatton Applications

13

1990 Final Market Share Estimates

Electronic Design Autofflation

All Platforms

Asia

Millions of Dollars/Actual Units

Company

Mentor Graphics

Zuken

Fujitsu

Cadence

NEC

Sun

Seiko Instruments--NO OEM

Valid

Uchida Yoko

Wacom

Sharp System Products--NO OEM

IBM

Hewlett-Packard

Racal-Redac

Digital

CADIX

Dazix

Toshiba—NO OEM

Hitachi

Synopsys

Scientific Calc.

Zycad

Tokyo Electron--NO OEM

Silvar-Lisco

Computervision

Viewlogic Systems

Data I/O

CAD AM

Sony

Sophia Systems

Compaq

EEsof

Ikos Systems

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped ss=====

97.8

=======

28.4

=======

41.1

73.5 32.9 40.6

=======

970

519

67.2

63.0

58.2

55.8

36.0

34.3

32.7

31.1

27.8

27.5

25.8

20.7

20.3

18.9

16.4

11.0

45.0

.0

39.2

48.7

15.5

19.6

31.1

6.2

14.5

20.6

19.2

1.3

14.8

7.6

3.2

5.5

15.5

63.0

14.4

.0

16.4

10.3

1.6

21.7

13.3

3.5

2.2

16.3

.0

9.4

1,492

0

3,581

3,613

230

247

519

509

166

2,795

1,633

101

175

139

121

97

10.4

7.9

6.8

6.1

5.9

5.1

4.9

4,4

4.3

4.0

4.0

3.5

3.5

2.9

2.7

5.0

.0

1.0

5.0

2.0

.0

2.2

.0

.0

1.8

3.5

,9

3.5

.0

2.7

6.5

4.4

4.4

7.1

4.0

.0

2.9

3.6

2.5

3.7

4.3

1.9

.0

2.3

.0

2.5

.0

320

0

24

50

21

0

51

0

0

250

230

50

581

2

12

===s==a

11.6% sssssss

7.0%

=======

12.0%

=======

4,6%

8.7%

7.9X

7.4%

6.9%

6.6%

4.3%

4.1%

3.9%

3.7%

3.3%

3.3%

3.0%

2.4%

2.4%

2.2%

1.9%

1.3%

1.2%

.9%

.8%

.7%

.7%

.6%

.6%

.5%

.5%

.5%

.5%

.4%

.4%

.3%

.3%

8.1%

11.1%

.0%

9.7%

12.1%

3.8%

4.8%

7.7%

1.5%

3.6%

5.1%

4.8%

.3%

3.7%

1.9%

.8%

1.4%

1.2%

.0%

.2%

1.2%

.5%

.0%

.5%

.0%

.0%

.4%

.9%

.2%

.9%

.0%

.7%

11.8%

4.5%

18.3%

4.2%

.0%

4.8%

3.0%

.5%

6.3%

3.9%

1.0%

.6%

4.8%

.0%

2.7%

1.9%

1.3%

1.3%

2.1%

1.2%

.0%

.8%

1.0%

.7%

1.1%

1,3%

.6%

.0%

.7%

.0%

.7%

.0%

2,4%

7,0%

.0%

16,9%

17.0%

1.1%

1.2%

2.5%

2.4%

.8%

13.2%

7.7%

.5%

.8%

.7%

.6%

.5%

1.5%

.0%

.1%

.2%

.1%

.0%

.2%

.0%

.0%

1.2%

1.1%

.2%

2.7%

.0%

.1%

41

(Continued)

®1991 Dataquest Incorporated June—Sepfoduction Ptohjbited

42

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Appllcatloas

13

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Asia

Millions of Dollars/Actual Units

Company

Century Research Center

Calay

Xilinx

Autodesk

Intergraph

Apple Computer

VLSI Technology

Seattle Silicon

LSI Logic

Meta-Software

CAD Software

Omron

Quickturn Systems

Test Systems Strategies

Logic Automation

Siemens

Altera

Microsim

Comdisco Systems

Analogy

Integrated Silicon Systems

Teradyne

BETRONEX

Quantic Laboratories

LPKF

Andor

EPIC Design Technology

Scientific & Engineering SW

SIMUCAD

Vantage Analysis Systems

ALDEC

Everex Systems

i i

©1991 Dataquest Incorporated June—Reproduction Prohibited

(Continued)

i

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share EstimatesElectronic D e s l ^ Automation Applications

13 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Asia

Millions of Dollars/Actual Units

Company

ACTEL

Motorola

Object Design

Kubota Computer

Tatvier Research

Accel Technologies

CAD Language Systems

NCR Microelectronics

Apt OS

Massteck

Omation

The CAD Group

Objectivity

Foresight Resources

Imagine That

National Semiconductor

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

' ~ ~ 1

.1

• 1

• 1

.1

. -1

.1

.1

.0

.0

.0

.0

.0

.0

.0

.0

15.3

846.7

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.1

.0

.0

.0

.0

14.8

404.3

.0

.0

.0

.0

.0

.0

.0

.0

.1

.1

.1

.0

.0

.0

.0

.0

.0

343.7

2,095

21,192

0

0

0

0

0

0

0

0

0

0

0

10

0

0

0

0

.0%

.0%

.0%

.OX

.OX

.OX

.OX

.OX

.0%

.0%

.0%

.OX

.0%

.0%

.0%

.0%

1.8X

100.OX

.OX

.OX

•OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.OX

.ox

.ox

.ox

.ox

.ox

3,7X

100.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.OX

.OX

.OX

.OX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

100.ox

9.9X

100.0X

436.5

384.6

189.9

211.3

176.3

148.4

13,072

7,942

25.6 3.0 19.0 178

133.6 116.6

.0

11,356

713.2 287.7 343.7 9,835

51.6X

45.4X

3.OX

47. ox

52.3X

.7X

51.3X

43.2X

5.5X

61.7X

37.5X

.8X

15.8X

28.8X .OX 53.6X

84.2X 71.2X 100.OX 46.4X

Source; Dataquest

June 1991

43

01991 Dataquest Iixorporated June—Reproduction Prohibited

44

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Mafket Share Esdmates—Ekctronk: Design Automation A|>p]lcatloas

14

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

Asia

Millions of Dollars/Actual Units

Company

rs:=====

Mentor Graphics

Zuken

Cadence

Sun

NEC

Valid

Uchida Yoko

Seiko Instruments--NO OEM

Fujitsu

Sharp System Products--NO OEM

Hewlett-Packard

Racal-Redac

CADIX

Dazix

Toshiba--NO OEM

Tokyo Electron--NO OEM

Scientific Calc.

Digital

Hitachi

Computervision

Zycad

Silvar-Lisco

Sony

Synopsys

Century Research Center

Calay

Ikos Systems

Wacom

Viewlogic Systems

Seattle Silicon

Intergraph

VLSI Technology

Omron

Total

Revenue

=======

Hardware Software

Revenue

=======

Revenue

=======

4.6

4.0

3.8

2.7

2.6

2.5

2.5

1.9

1.7

1.6

1.5

1.4

97.8

69.8

54.8

48.9

35.3

34.3

32.7

32.1

27.6

24.2

22.2

20.7

18.9

16.4

11.0

5.9

5.7

5.3

5.2

4.9

4.9

28.4

31.1

.0

43.0

19.7

19.6

31.1

13.8

18.5

12.6

16.4

1.3

7.6

3.2

5.5

2.0

.7

3.8

2.5

2.2

3.8

.0

3,5

.0

1.4

.8

2.5

.6

.0

.0

.9

.1

1.2

41.1

38.7

54.8

.0

12.7

10.3

1.6

15.1

6.4

11.6

2.2

16.3

9.4

6.5

4.4

Wkstns

Shipped

=======

970

486

0

3,307

1,853

247

519

215

649

150

1,185

101

139

121

97

21

18

175

108

51

41

0

230

0

15

43

11

1.0

1.5

.0

1.9

1.6

1.2

.3

1.2

.1

2.9

3.1

.0

2.2

2.5

.0

3.2

.0

3.4

25

0

0

17

4

39

- Market Share -•

Total i ffardware Software

Revenue Revenue Revenue

Wkstns

Shipped

=======

=======

======= ======

15.6%

11.2%

8.8%

7.8%

5.6%

5.5%

5.2%

5.1%

4.4%

3.9%

3.6%

3.3%

3.0%

2.6%

1.8%

.9%

.9%

.8%

.8%

.8%

.8%

.7%

.6%

.6%

.4%

.4%

.4%

.4%

.3%

.3%

.3%

.2%

10.1%

11.1%

.0%

15.3%

7.0%

7.0%

11.1%

4.9%

6.6%

4.5%

5.8%

.5%

2.7%

1.1%

2.0%

.7%

.3%

1.4%

.9%

.8%

1.4%

.0%

1.2%

.0%

.5%

.3%

.9%

.2%

.0%

.0%

.3%

.0%

15.3%

14.4%

20.4%

.0%

4.7%

3.8%

.6%

5.6%

2.4%

4.3%

.8%

6.1%

3.5%

2.4%

1.6%

1.1%

1.2%

.0%

.8%

.9%

.0%

1.2%

.0%

1.3%

.4%

.6%

.0%

.7%

.6%

.4%

.1%

.5%

8.9%

4.5%

.0%

30.3%

17.0%

2.3%

4.8%

2.0%

5.9%

1.4%

10.9%

.9%

1.3%

1.1%

.9%

.2%

.2%

1.6%

1.0%

.5%

.4%

.0%

2.1%

.0%

.1%

.4%

.1%

.2%

.0%

.0%

.2%

.0%

.2% .4% .1% .4%

i i

(Continued)

i

01991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share EstimatesElectronic D e s i ^ Automation Applications

14

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

Asia

Millions of Dollars/Actual Units

Company

LSI Logic

Quickturn Systems

EEsof

Sophia Systems

Logic Automation

Meta-Software

Siemens

Test Systems Strategies

IBM

Comdisco Systems

Analogy

Data I/O

Integrated Silicon Systems

Quantic Laboratories

Teradyne

EPIC Design Technology

Scientific & Engineering SW

SIMUCAD

Autodesk

Vantage Analysis Systems

Xilinx

Microsim

Motorola

Object Design

CAD Language Systems

NCR Microelectronics

Kubota Computer

Objectivity

ACTEL

Everex Systems

National Semiconductor

.6

.3

.3

.3

.2

.2

.2

.2

.1

.1

1.3

1.3

1,3

1.3

1.2

1.2

1.2

1.1

1.0

.9

.8

.7

.6

.1

.1

.1

.0

.0

. 0 •

.0

.0

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

0

0

0

0

0

0

22

0

0

0

2

0

3

10

2

10

0

0

15

0

0

0

0

0

0

0

7

0

0

1

0

.6

.0

.0

.0

.1

.0

.0

.0

.1

1.3

.0

.3

.0

.0

.6

.0

,0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.5

.3

.3

.3

.2

.2

.2

.2

.8

.7

.7

.5

1.0

.0

1.1

.8

1.2

1.1

.4

.9

.2

.1

.1

.1

.0

.0

.0

.0

.0

.0

.0

.2%

.1%

.1%

.1%

.1%

.1%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.2%

.n

.0%

.0%

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.IX

.ox

.ox

.IX

.ox

.2X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.5%

.ox

.IX

.ox

.ox

.2X

.ox

.2X

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.IX

.IX

.IX

.IX

.IX

.IX

.OX

.OX

.OX

.4X

.ox

.4X

.3X

.4X

.4X

.IX

.3X

.IX

.3X

.3X

.2X

.2X

.2X

.ox

.OX

.ox

.ox

.ox

.ox

.ox

45

(Continued)

&1991 Dataquest Incorporated June—Reproduction Prohibited

46

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electranic Design Automatlan Applications

14 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

Asia

Millions of Dollars/Actual Units

Company

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Coofianies

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.0

625.3

.0

280.5

.0

269.0

0

10,909

.0%

100.0%

.0%

100.0%

.0%

100.0%

.0%

100.0%

326.4

274.4

24.5

73.8

551.5

126.6

151.2

2.7

64.0

216.5

142.0

108.8

18.3

.0

269.0

6,188

4,563

158

4,669

6,240

52.2%

43.9%

3.9%

11.8%

88.2%

45.1%

53.9%

1.0%

22.8%

77.2%

52.8%

40.4%

6.8%

.0%

100.0%

56.7%

41.8%

1.5%

42.8%

57.2%

Source: Dataquest

June 1991

i i

01991 Dataquest bKorpoiated June—Reproduction Prohibited

i

TABLE NUNBER:

TITLE:

APPLICATION:

P U T FORM:

REGION:

UNITS:

Market Share EstimatesElectronic D e s ^ Automatiott Applications

15

1990 Final Market Share Estimates

Electronic Design Automation

Host/Server

Asia

Millions of Dollars/Actual Units

Company

— — —

Fuj i tsu

Digital

IBM

Cadence

Sun

Synopsys

Seiko Instrunents-'NO OEM

Zuken

Sharp System Products--NO OEM

Hitachi

Zycad

Scientific Calc.

Hewlett-Packard

Intergraph

Silvar-Lisco

Teradyne

CAD AM

Data I/O

Meta-Software

VLSI Technology

LSI Logic

EEsof

Test Systems Strategies

Analogy

Kubota Computer

Microsim

SIMUCAD

Logic Automation

Total

Revenue

Hardware Software

Revenue Revenue

Ukstns

Shipped

=ssss=ss

28.9

—-"-—=*

19.4

15.0

15.0

8.2

6.9

11.0

9.0

.0

5.7 sssss==

6.7

=======

489

.0

3.3

0

174

8.2

.0

0

306

>-1

3.9

3.7

3.6

1.7

1.2

1.1

1.0

.0

1.7

1.8

1.9

.8

1.1

.2

.8

3.7

1.3

1.9

1.7

.7

.0

.9

.0

0

15

33

16

138

9

6

51

.5

.5

.3

.2

.2

.2

.2

.3

.0

.0

.0

.0

.0

.0

.1

.4

.2

.2

.2

.2

.2

5

0

0

0

0

0

1

0

.2

.1

.1

.0

.0

.0

.0

.0

.0

.1

.1

.1

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

0

0

0

3

0

0

0

Total 1

Revemje

- Market

Revenue

Software

Revenue

Wkstns

Shipped

— • — —

27.4%

14.2%

14.2%

7,7%

6.5%

3.9%

==sss=s

31.4%

=======

22.0%

=======

30.9%

17.8%

.0% .0%

14.5%

.0%

9.2%

.0%

10.9%

26.9%

.0%

12.2%

11.0%

.0%

19.3%

.0%

3.7%

3.5%

3.4%

1.6%

1.2%

1.0%

.9%

.5%

.5%

.3%

.2%

.2%

.2%

.2%

.1%

.1%

.1%

.0%

.0%

.0%

.0%

.0%

2.7%

2.9%

3.0%

1.3%

1.8%

.4%

1.3%

.4%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

4.4%

6.4%

5.7%

2.3%

.0%

2.8%

.0%

.2%

1.2%

.8%

.7%

.7%

.7%

.5%

.4%

.4%

.4%

.1%

.1%

.1%

.1%

.1%

1.0%

2.1%

1.0%

.0%

.0%

.0%

.0%

.2%

.0%

.0%

.0%

8.7%

.6%

.3%

3.2%

.3%

.0%

.0%

.0%

.0%

.0%

.0%

47

(Continued)

01991 Dataquest Incorporated June—Reproduction Prohibited

48

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Antomatloa Applications

15 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Host/Server

Asia

Millions of Dollars/Actual Units

Company

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

8.6

105.5

8.1

61.8

.0

30.3

338

1,584

8.1%

100.0%

13.1%

100.0%

.0%

100.0%

21.3%

100.0%

63.7

41.8

. .0

31.5

74.0

36.2

25.5

.0

25.6

36.2

17.9

12.4

.0

.0

30.3

889

695

0

695

889

60.4%

39.6%

.0%

29.8%

70.2%

58.7%

41.3%

.0%

41.4%

58.6%

59.2%

40.8%

.0%

.0%

100.0%

56.1%

43.9%

.0%

43.9%

56.1%

Source: Dataquest

i i

©1991 Dataquest Incojporated June—Reproduction Prohibited

i

TABLE NUHBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Maricet Share Estimates—Electronic Design Aatomatloa AppJicatlans

16

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

Asia

Millions of Dollars/Actual Units

Company

Uacorn

NEC

IBM

Fujitsu

CADAM

Hitachi

Compaq

Data I/O

Hewlett-Packard

Viewlogic Systems

Xilinx

Sophia Systems

Autodesk

Apple Computer

EEsof

CAD Software

Altera

Microsira

BETRONEX

LPKF

Andor

Integrated Silicon Systems

Ikos Systems

Teradyne

ALDEC

Everex Systems

ACTEL

Tanner Research

Accel Technologies

Apt OS

Massteck

Omation

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

28.6

.1

.1

.1

.1

.0

.0

.0

.5

.3

.2

.2

.1

.1

22.9

11.6

10.8

3.8

3.5

3.5

3.5

2.6

2.5

2.3

2.2

2.1

1.9

1.4

1.4

1.1

.8

.7

484

1,728

2,600

354

250

74

581

0

397

0

37

0

0

3

3

1

0

0

0

0

0

• 0

0

40

0

370

0

0

0

0

0

20

.1

.2

.1

.0

.1

.1

.0

.1

.0

2.1

2.1

1.5

2.1

.0

1.3

1.2

.9

.8

.7

.0

.0

.0

.0

19.9

1.7

.0

2.5

1.7

1.5

.0

3.5

.0

5.6

19.4

11.1

7.2

1.8

1.7

3.5

.0

2.0

.0

.0

.6

.0

1.7

.0

.0

.0

.0

.0

.3

.1

.1

.2

.0

.0

.1

.0

.0

.0

.0

.0

.0

24.7%

19.8%

10.0%

9.3%

3.3%

3.1%

3.0%

3.0%

2.2%

2.2%

2.0%

1.9%

1.8%

1,7%

1.2%

1.2%

.9%

.7%

.6%

.4%

.3%

.2%

.2%

.1%

.1%

.1%

.1%

.1%

.0%

.0%

.0%

.0%

.1%

.1%

.3%

.0%

.0%

.2%

.0%

.0%

.0%

.0%

.0%

.0%

9.1%

31.3%

17.8%

11.6%

2.8%

2.7%

5.6%

.0%

3.3%

.0%

.0%

.9%

.0%

2.7%

.0%

.0%

.0%

.0%

.0%

.5%

44.7%

3.8%

.0%

5.6%

3.9%

3.4%

.0%

7.8%

.0%

4.8%

4.6%

.3%

.5%

.3%

.0%

.2%

.3%

.0%

.2%

.1%

.1%

.1%

.1%

.1%

3.3%

4.7%

.0%

2.9%

2.6%

2.1%

1.8%

1.5%

5.6%

19.9%

29.9%

4.1%

2.9%

.8%

6.7%

.0%

4.6%

.0%

.0%

.5%

.0%

4.2%

.0%

.0%

.0%

.0%

.0%

.2%

.0%

.0%

.0%

.0%

.0%

.4%

.0%

.0%

.0%

.0%

.0%

.0%

49

(Continued)

01991 Dataquest Incorporated June—Reproduction Prohibited

50

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estbnates—Electronic D e s ^ Automatloa Appllcatlotis

16 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

Asia

Millions of Dollars/Actual Units

i

Company

The CAD Group

Foresight Resources

Imagine That

Meta-Softuare

SIMUCAD

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Companies

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.0

.0

.0

.0

.0

6.7

116.0

.0

.0

.0

.0

.0

6.7

62.0

.0

.0

.0

.0

.0

.0

44.4

0

0

0

0

0

1,757

8,698

.0%

.0%

.0%

.OX

.OX

5.8X

100.0X

.OX

.OX

.OX

.OX

.OX

10.8X

100.0X

.IX

.OX

.OX

.OX

.OX

.OX

100.0X

.OX

.OX

.OX

.OX

.OX

20.2X

100.0X

46.4

68.4

1.1

28.4

87.6

27.1

34.6

.3

27.0

35,0

16.4

27.2

.8

.0

44.4

5,995

2,683

20

5,992

2,706

40.OX

59. OX

1.0X

24.5X

75.5X

43.7X

55.8X

.5X

43.6X

56.4X

36.9X

61.3X

1.7X

.OX

100.0X

68.9X

30.8X

.2X

68.9X

31. IX

Source: Dataquest

June 1991

i

01991 Dataquest Incoipotated June—Repioduoioa Prohibited

(

Company

=======

Digital

Hewlett-Packard

IBM

Compaq

Sun

Oread

Genrad

LSI Logic

Solbourne

Computervision

Data I/O

Intergraph

Altera

Analogy

VLSI Technology

Xilinx

Royal Digital Systems

EEsof

Siemens

Zuken

Teradyne

Apple Computer

Autodesk

BETRONEX

LPKF

ACTEL

CAD Software

Spectrum Software

Accel Technologies

Capilano Computing

Massteck

Ziegler

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Esrimaiw Wcctronte Ve^ga Antomattoa Apidications

17

1990 Final Market Share Estimates

Electronic Design Automation

AllPlatfomis

Rest of World

Millions of Dollars/Actual Units

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped s=assss

6.8

=======

4.9

=======

.0

======

58

3.9

3.7

3.1

3.1

.2

.3

286

538

3.5

2.8

1.8

.8

3.5

2.4

.0

.2

.0

.0

1.8

.6

581

153

0

9

.7

.6

.6

.5

.5

.5

.4

.4

.3

.3

.3

.2

.2

.2

.2

.2

.1

.6

.2

.0

.3

.0

.0

.0

.0

.0

.0

.1

.1

.0

.2

.0

.6

.0

.1

.5

.1

.4

.4

.3

.3

.2

.2

.1

.1

.2

.0

2

45

4

0

5

0

0

3

2

0

34

0

1

0

0

0

0 .1

.1

.1

.1

.0

.1

.0

.0

.2

.1

.0

.1

.1

.1

.0

.0

.1

.1

.0

5

0

0

0

.0

.0

.0

.0

.0

.0

.0

.0

.0

0

0

0

0

=======

19.5%

=======

21.4% sssssss

.0% sssssss

2.9%

11.3X

10.8%

10.1%

8.0%

5.3%

2.4%

2.1%

1.8%

1.6%

1.6%

1.5%

1.4%

1.2%

1.1%

1.0%

.8%

.8%

.7%

.6%

.5%

.5%

.5%

.3%

.3%

.3%

.3%

.3%

.1%

.1%

.1%

.1%

13.4%

13.3%

15.2%

10.5%

.0%

.7%

.3%

2.6%

.7%

.0%

1.1%

.0%

.0%

.0%

.0%

.0%

.0%

.5%

.3%

.0%

.7%

.0%

.0%

.3%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

2.9%

4.3%

.0%

.0%

25.5%

7.6%

7.6%

.0%

1.9%

7.5%

1.2%

6.0%

5.1%

4.2%

4.0%

3.3%

3.1%

1.1%

1.7%

2.2%

.0%

2.5%

1.7%

.4%

1.2%

1.2%

1.4%

.6%

.6%

.6%

.4%

14.2%

26.7%

28.8%

7.6%

.0%

.4%

.1%

2.2%

.2%

.0%

.3%

.0%

.0%

.1%

.0%

.0%

.0%

.2%

.1%

.0%

1.7%

.0%

.0%

.2%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

51

(Continued)

01991 Daaquest Incorpotated June Reproduaton Prohibited

52

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design AutomatlcMi Appllcatioas

17 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

All Platforms

Rest of World

Millions of Dollars/Actual Units

Company

American Small Business Coinp.

Other Companies

All Conpanies

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

.0

4.5

34.6

.0

4.3

23.0

.0

.0

7.2

0

290

2,016

.0%

13. OX

100.0X

.0%

18.7%

100.OX

.1%

.OX

100.0X

.OX

14.4X

100.OX

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SU Companies

33.9

.2

.5

23.6

11.1

22.7

.1

.2

20.6

2.4

6.8

.1

.3

.0

7.2

2,007

2

8

1,944

72

97.9%

.6%

1.4X

68.0X

32. OX

98.9X

.3X

.8X

89.6X

10.4X

94.7X

1.7X

3.6X

.OX

100.OX

99.5%

.IX

.4X

96.4X

3.6X

Source: Dataquest

June 1991

i

(

01991 Daiaquest Incorporated June—ReproductiCQ Prohibited

i

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

MariEct Share EsOnaxts—Electroolc Design Antiwnatton AppUcatlofis

18

1990 Final Market Share Estimates

Electronic Design Automation

Technical Workstation

Rest of World

Millions of Dollars/Actual Units

Conpany

Hewlett-Packard

Sun

Digital

Genrad

LSI Logic

Computervision

Analogy

Intergraph

Solbourne

VLSI Technology

Royal Digital Systems

Siemens

Zuken

EEsof

Data I/O

IBM

Teradyne

Xilinx

ACTEL

Autodesk

Other Companies

All Companies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SW Conpanies

2.7

2.1

1.3

.1

.1

.2

.0

.2

.3

.0

.0

.1

.1

.0

.0

.1

.0

.0

.0

.0

.0

7.2

7.0

.1

.1

5.7

1.4

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.2

.1

.1

.1

.1

.0

.0

3.4

2.4

1.8

.7

.7

.6

..4

.4

.3

.3

.3

.2

.0

.0

12.0

.2

.0

.0

.5

.5

.1

.4

.1

.0

.3

.2

.1

.1

.1

.1

.0

.1

.0

.0

.0

.0

2.7

231

140

58

6

2

4

0

4

23

1

0

3

1

0

0

2

0

0

0

0

0

475

48.5X

29.4X

12.3X

1.4X

.4X

.8X

.OX

.9X

4.8X

.2X

.OX

.6X

.3X

.OX

.OX

.5X

.OX

.OX

.OX

.OX

.OX

100.0X

7.7X

.OX

.OX

17.2X

17.9X

5.IX

12.8X

2.6X

.OX

9.9X

8.8X

2.9X

4.0X

3.3X

2.9X

.7X

2.6X

.7X

.4X

.4X

.OX

100.0X

36.9X

29.8X

17.5X

1.9X

1.0X

2.2X

.OX

2.9X

4.2X

.IX

.OX

1.5X

1.1X

.OX

.OX

.7X

.OX

.OX

.OX

.OX

.OX

100.0X

1.0X

.7X

.7X

.7X

.2X

.2X

.IX

.OX

100.0X

28. IX

20.3X

14.6X

6.0X

5.4X

4.7X

3.2X

3.2X

2.7X

2.7X

2.2X

1.9X

1.7X

11.6

.2

.2

6.9

5.1

2.5

.1

.1

.0

2.7

471

1

3

428

48

96.4X

1.7X

1.9X

57.4X

42.6X

97.4X

1.1X

1.5X

79.9X

20.1X

93.0X

4.0X

2.9X

.OX

100.0X

99.1X

.3X

.6X

90.0X

10.0X

Source: Dataquest

June 1991

53

01991 Dataquesc Incorporated June—Reproduction Prohibited

54

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estiiaates—Electroalc Design Antomatioa Applications

19

1990 Final Market Share Estimates

Electronic Design Automation

Host/Server

Rest of World

Millions of Dollars/Actual Units

Conpany

Digital

IBM

Sun

SoIbourne

Hewlett-Packard

Intergraph

Genrad

Teradyne

LSI Logic

VLSI Technology

Data I/O

Analogy

EEsof

Zuken

Other Companies

All Connpanies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companies

All Hardware Companies

All Turnkey & SU Companies

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

.1

,1

.1

• 1

.0

.0

5.0

1.3

.3

.3

.2

.1

.0

.0

4.0

11.7

11.7

.0

.0

9.9

1.8

3.7

.8

.3

.3

.2

.1

.0

.0

.0

.0

.0

.0

.0

.0

3.8

9.1

9.1

.0

.0

8.2

.9

.6

.0

•0

.0

.6 it)'

.1

.:i

.1

.0^

^

*0

^0

J

.O'

.0

.0

•0

.0

.0

-*

0

16

13

22

12

1

1

0

0

0

0

0

0

0

158

224

224

0

0

206

18

42.9%

11.4%

2.9%

2.5%

2.0%

1.1%

.7%

.7%

.6%

.4%

.3%

.2%

.1%

.1%

34.2%

100.0%

99.9%

.1%

.0%

84.5%

15.5%

40.2%

8.8%

3.1%

3.3%

2.1%

.5%

.2%

.0%

.1%

.0%

.0%

.0%

.0%

.0%

41.6%

100.0%

.0%

49.2%

.0%

.0%

.0%

3.4%

10.2%

10.2%

10.2%

5.1%

5.1%

3.4%

1.7%

1.7%

.0%

100.0%

100.0%

.0%

.0%

90.3%

9.7%

.0%

7.0%

6.0%

10.0%

5.2%

.5%

.4%

.0%

.1%

.0%

.0%

.0%

.0%

.0%

70.7%

100.0%

98.3%

1.7%

.0%

.0%

100.0%

100.0%

.0%

.0%

91.9%

8.1%

Source: Dataquest

June 1991

01991 Dataquest Incorporated June-^eptoducticm I>rohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Design Automation Applications

20

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

Rest of World

Millions of Dollars/Actual Units

Company

Compaq

IBM

Oread

Altera

Data I/O

Xilinx

Hewlett-Packard

Apple Computer

Autodesk

EEsof

BETRONEX

LPKF

CAD Software

Spectrum Software

ACTEL

Accel Technologies

Capilano Computing

Massteck

Teradyne

Ziegler

Genrad

American Small Business Comp.

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

= = = = x =

3.5

======= ======= =======

3.5 .0 581

2.3

1.8

2.2

.0

.0

1.8

520

0

.5

.4

.3

.3

.2

.2

.1

,1

• 1

.1

.1

.1

.0

.0

.0

.2

.2

.0

.0

.0

.1

.0

.0

.0

.4

.4

.3

.0

.0

.2

.1

.1

.0

.1

.1

.1

0

0

0

44

34

0

0

0

S

0

0

0

.1

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

.0

0

0

0

0

0

1

0

=:=ss=s=

=======

32.0% 52.3%

====s==

===s=s:s

.0%

44.1%

21.3%

16.9%

4.6%

3.9%

2.7%

2.7%

33.1%

.0%

.0%

.0%

.0%

3.6%

.0%

47.3%

11.1%

11.1%

6.9%

.0%

39.5%

.0%

.0%

.0%

.0%

3.4%

1.6%

1.6%

1.2%

1.1%

1.1%

2.4%

.0%

.0%

.0%

1.0%

.0%

4.4%

3.1%

3.1%

.8%

2.6%

.0%

.0%

.0%

1.0%

.9%

.8%

.5%

.0%

.0%

.0%

.0%

2.3%

2.6%

2.1%

1.0%

.4%

.0%

.0%

.0%

.4%

.4%

.3%

.3%

.2%

.1%

.0%

.0%

.0%

.0%

.0%

.0%

1.0%

1.0%

.8%

.8%

.5%

.3%

.0%

.0%

.0%

.0%

.0%

.1%

.0%

55

01991 Daiaquest Incorpoiaced June—Reproduction Prohibited

(Continued)

56

TABLE NUMBER:

T I T L E :

APPLICATION:

PLATFORM:

REGION:

UNITS:

Marlcet Share Estimates—Electronic Design Antomatian Applications

20 (Continued)

1990 Final Market Share Estimates

Electronic Design Automation

Personal Computer

Rest of World

Millions of Dollars/Actual Units

Company

Other Companies

All Conpanies

All U.S.-Based Companies

All Asian-Based Companies

All European-Based Companie

All Hardware Companies

All Turnkey & SU Companies

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

.5

10.9

.5

6.7

.0

3.9

132

1,317

4.6%

100.0%

7.5%

100.0%

.0%

100.0%

10.0%

100.0%

10.6

.0

. .3

6.8

4.1

6.6

.0

.1

6.6

.1

3.7

.0

.2

.0

3 . 9

1,312

0

5

1,311

6

97.5%

.0%

2.5%

62.1%

37.9%

99.0%

.0%

1.0%

99.0%

1.0%

95.4%

.0%

4.6%

.0%

100.0%

99.6%

.0%

.4%

99.5%

.5%

Source: Dataquest

June 1991

i i

®1991 Dataquest Incorporated June—Reproductioa Prohibited

i

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic Des^a Antamatioa Applications

21

1990 Final Market Share Estimates

Electronic CAE

A U Platforms

Worldwide

Millions of Dollars/Actual Units

Company

Mentor Graphics

Hewlett-Packard

Sun

Digital

Valid

Cadence

Compaq

Dazix

IBM

NEC

Wacom

Zycad

Racal-Redac

Fujitsu

Viewlogic Systems

Intergraph

Synopsys

EEsof

Siemens

Teradyne

Computervision

Autodesk

Xilinx

Ikos Systems

LSI Logic

Zuken

Logic Automation

VLSI Technology

Data I/O

ABB Cade International

Aucotec

ACTEL

Altera

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total

- narKei snare -•

Hardware Software

Wkstns

Revenue

Revenue Revenue Shipped

34.2

31.1

30.4

27.8

27.7

27.5

26.5

26.2

26.0

20.5

18.8

18.8

16.7

16.5

249.1

151.5

145.6

92.6

91.9

• 86.3

82.0

64.9

39.7

13.5

13.0

12.6

11.8

11.1

10.8

10.6

10.5

10.0

10.0

22.8

6.2

24.9

1.5

17.2

.0

13.5

1.3

5.6

.0

.4

.0

6.4

3.7

.0

.0

14.2

.0

.1

10.3

.0

6.9

.0

.0

72.2

115.9

127.0

65.7

20.2

.0

82.0

13.0

37.7

104.6

9.9

.0

2.0

52.3

68.3

.0

26.0

.0

8.7

21.7

.0

22.5

7.8

23.1

4.6

23.6

22.7

7.2

15.5

4.6

16.7

14.8

.0

9.8

7.0

11.8

9.2

10.8

3.2

5.8

9.0

8.5

2,468

10,398

7,707

776

549

0

13,665

478

8,864

2,088

509

250

125

614

0

273

0

18

272

0

145

0

0

61

33

88

0

32

0

46

167

0

0

14.8%

9,0%

8.6%

5.5X

5.4%

5.1%

4.9%

3.8%

2.3%

2.0%

1.8%

1.8%

1.0%

1.0%

.8%

.8%

.7%

.7%

.7%

.6%

.6%

.6%

.6%

.6%

1.6%

1.6%

1.6%

1.6%

1.6%

1.5%

1.2%

1.1%

1.1%

.0%

.1%

.0%

.8%

.5%

.0%

.0%

.0%

.0%

1.3%

.0%

.9%

.0%

.0%

1.8%

.2%

.7%

9.5%

15.2%

16.7%

8.6%

2.7%

.0%

10.8%

1.7%

4.9%

3.0%

.8%

3.3%

.2%

2.3%

.0%

1.9%

.0%

.0%

.4%

.0%

.2%

.0%

.0%

3.4%

.8%

.4%

.2%

1.0%

.0%

.4%

4.0%

16.8%

12.5%

1.3%

.9%

.0%

22.1%

.8%

14.4%

.1%

.1%

.1%

.0%

.1%

.0%

.1%

«3%

.0%

.0%

2.3%

.0%

1.5%

1.1%

1.8%

1.4%

1.7%

.5%

.9%

1.4%

1.3%

16.2%

1.5%

.0%

.3%

8.1%

10.6%

.0%

4.0%

.0%

1.3%

3.4%

.0%

3.5%

1.2%

3.6%

.7%

3.6%

3.5%

1.1%

2.4%

.7%

2.6%

57

(Continued)

01991 Dataquesc Incorponued June—^teproductlon Prohibited

58

TABLE NUMBER:

T I T L E :

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Esdmaies—Etectronlc Design Automation Applications

21

1990 Final Market Share Estimates

Electronic CAE

All Platforms

Worldwide

Millions of Dollars/Actual Units

Company

Apple Computer

Genrad

Analogy

Microsim

Kloeckner-Moeller

Oread

Quickturn Systems

Meta-Software

Comdisco Systems

Ascent Logic Corp

Test Systems Strategies

Calay

Vantage Analysis Systems

AnaCAO

Ziegler

Everex Systems

Quad Design Technology

EPIC Design Technology

SoIbourne

Motorola

Quantic Laboratories

Sophia Systems

Assigraph

Tokyo Electron--NO OEM

Scientific & Engineering SW debis Systemhaus

ALDEC

ALS Design

Scientific Calc.

SIMUCAD

CAD/CAM Group

Research Machines

Phase Three Logic

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

=======

8.7

=======

7.7

8.3

8.1

1.6

.0

7.5

6.8

6.5

6.5

6.2

6.2

6.0

6.0

5.2

4.0

3.9

3.7

3.6

3.1

3.0

3.0

2.9

2.9

2.8

2.8

2.8

2.7

2.6

2.5

1.9

1.9

1.8

1.7

.0

4.1

.0

6.5

.0

.0

.0

.0

1.7

.0

.0

.0

3.6

.0

.0

3.0

.0

.0

.7

.0

.9

.0

.7

.0

.0

.3

.0

.0

=======

.0

=======

1,680

5.5

7.3

7.1

2.4

6.5

.0

6.1

5.6

4.8

5.2

3.1

4.0

3.9

3.7

.0

2.9

2.7

.0

2.9

2.7

1.8

2.2

1.3

2.7

1.5

2.5

1.9

1.2

1.7

1.7

86

0

0

107

0

50

0

0

0

0

85

0

0

0

1,087

0

0

224

0

0

43

21

10-

0

90

0

0

8

0

0

1.6

1.5

1.6

.0

.0

1.3

242

0

=======

.5%

.5%

.5%

.4%

.4%

.3%

.2%

.2%

.2%

.2%

.2%

.2%

.4%

.4%

.4%

.4%

.4%

.4%

.2%

.2%

.2X

.2%

.2%

.2%

.2%

.2%

.IX

.1%

.1%

.IX

.IX

.1%

.1X

=======

1.0%

.2X sssssss

.ox

.8X

=======

2.7X

.IX

.0%

.OX

.5X

.0%

.9X

• OX

.OX

1.1X

1.1X

.4X

1.0X

.ox

.9X

.0%

.0%

.2%

.OX

.9X

.7X

.8X

.5X

.6X

.6%

.6X

.OX

.OX

.2X

.OX

.IX

.OX

.OX

.OX

.OX

.IX

.OX

.OX

.OX

.OX

.OX

.5%

.OX

.OX

.4X

.OX

.OX

.IX

.OX

.1%

.OX

.IX

.OX

.ox

.OX

.ox

.0%

.2X

.OX

.ox

.4%

.4X

.OX

.4%

.4X

.3X

.3%

.2X

.4X

.2X

.4X

.3X

.2X

.3X

.3X

.OX

.2X

1.8X

.OX

.OX

.4X

.OX

.OX

.1%

.OX

.OX

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.4X

.ox

(Continued)

&1991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Eailmiifri Tlccnonir Deslga Automation AppUcations

21

1990 Final Market Share Estimates

Electronic CAE

All Platforms

Worldwide

Millions of Dollars/Actual Units

Company

Sony

Century Research Center

CAD Software

Omron

Caditron

Seattle Silicon

Omation

Dell Computer

Spectrin Software

National Semiconductor

Serbi

Aucos elektronische Gerate

Intrinsix rotring euroCAD

I SKA

Infinite Graphics

MacNeaI-SchwendIer

Intercad

DAT Standard Information system

Inca

Visionics

Ontologic

Douglas Electronics

CAD Language Systems

Capilano Computing

NCR Microelectronics

Accel Technologies

PLUS Logic

The CAD Group

Imagine That

Foresight Resources

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

Market Share

Total Hardware Software Wkstns

Revenue Revenue Revenue Shipped

1.4

1.4

1.4

1.4

1.3

1.2

1.2

1.1

1.1

1.0

1.0

1.0

1.0

1.0

.9

.8

.8

.8

.8

.7

.5

.4

.4

.4

.4

.7

.6

.6

.5

.5

.4

0

38

0

0

3

0

0

0

0

0

8

0

100

10

23

16

0

84

8

0

39

20

0

0

279

0

0

0

0

0

0

.0

.0

.4

.0

.0

.0

.0

.0

.0

.0

.0

.2

.0

.3

1.0

.5

.4

1.3

.8

.0

1.2

.6

.0

.0

1.1

.0

.0

.0

.0

.0

.0

.5

.6

.6

.4

.5

.8

.8

.2

.7

.7

1.1

.9

1.0

.7

.0

.4

.4

.0

.5

1.2

.1

.5

.8

1.2

.0

.4

.4

.4

.4

.4

.3

0%

ox

0%

0%

ox ox ox ox ox ox ox ox ox ox ox ox

1%

1%

1%

1%

1%

1%

1%

.1%

.1%

1%

1%

1%

1%

n

1%

.ox

.ox

.ox

.2X

.ox

.ox

.0%

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.IX

.ox

.ox

.IX

.ox

.ox

.ox

.5%

.IX

.2X

.IX

.ox

.IX

.IX

.IX

.IX

.ox

.IX

.IX

.ox

.IX

.2X

.ox

.IX

.IX

.2X

.ox

.2X

.IX

.IX

.IX

.IX

.IX

.IX

.IX

.IX

.IX

.1X

.ox

.2X

.IX

.ox

.2X

.IX

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.IX

.IX

.IX

.ox

.ox

.IX

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

.ox

59

(Continued)

®1991 Dataquest locorporated June—Reproduction Profaibtted

60

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Maricet Shate Estimates—Electronic Design Aiitomatlcm AppUcatlans

21 (Continued)

1990 Final Market Share Estimates

Electronic CAE

All Platforms

Worldwide

Millions of Dollars/Actual Units

Company

Market Share

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

Total Hardware Software Ukstns

Revenue Revenue Revenue Shipped

0

0

0

0

Objectivity .2

0

0

0

0

0

3

0

0

0

0

7,743

.0%

.0%

.0%

.0%

.0%

.0%

.0%

3.2%

100.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

6.9%

100.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

100.0%

.1%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

12.5%

100.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

.0%

All U.S.-Based Companies 1,466.7 672.4 535.9 56,895

All Asian-Based Companies 115.4 56.6 49.0 3,483

All European-Based Companies 105.6 32.0 62.2 1,356

All Hardware Companies 553.0 487.7 .0 51,472

All Turnkey & SW Companies 1,134.7 273.5 647.1 10,263

86.9% 88.3% 82.8% 92.2%

6.8% 7.4% 7.6% 5.6%

6.3% 4.2% 9.6% 2.2%

32.8% 64.1% .0% 83.4%

67.2% 35.9% 100.0% 16.6%

Source: Dataquest

June 1991

®1991 Dataquest Incorporated June—Reproduction Prohibited

TABLE NUMBER:

TITLE:

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share EstiBPatrs Flectronlc Design Amtomatlon Applications

22

1990 Final Market Share Estimates

Electronic CAE

Technical Workstation

Worldwide

Millions of Dollars/Actual Units

Company

Mentor Graphics

Sun

Hewlett-Packard

Valid

Cadence

Dazix

Digital

Zycad

Racal-Redac

NEC

Siemens

Computervision

Intergraph

Zuken

Synopsys

Ikos Systems

EEsof

LSI Logic

Viewlogic Systems

Logic Automation

Fuj i tsu

ABB Cade International

VLSI Technology

Teradyne

Analogy

Genrad

Quickturn Systems

Comdisco Systems

Ascent Logic Corp

Test Systems Strategies

Calay

Meta-Software

Vantage Analysis Systems

Total

Revenue

Hardware Software

Revenue Revenue

Wkstns

Shipped

- narKei anare -•

Wkstns

Revenue Revenue Revenue Shipped

249.1

123.8

120.0

18.4

16.9

12.6

12.6

12.6

11.7

11.7

11.6

91.9

79.2

.64.9

25.2

24.3

24.2

21.6

20.5

11.6

11.4

10.6

10.2

7.9

7.7

7.2

6.5

6.2

6.0

5.4

5.2

5.2

4.0

9.8

8.8

9.8

11.6

3.2

3.2

8.5

6.6

6.9

4.8

.0

5.6

4.8

4.6

3.1

5.2

4.0

104.6

.0

9.9

52.3

62.5

26.0

1.5

.0

19.1

7.8

7.2

4.5

3.1

7.0

11.3

.0

2,468

6,884

6,057

549

0

478

776

204

125

1,134

272

137

187

88

0

57

50

0

0

0

85

0

0

18

30

0

0

267

46

29

0

0

64

72.2

108.9

90.5

20.2

.0

13.0

17.0

19.2

1.5

12.1

10.3

6.8

9.5

5.6

.0

12.6

.1

1.2

.0

.0

7.0

6.4

.3

.0

.0

1.4

6.5

.0

.0

.0

1.7

.0

.0

1.0%

1.0%

1.0%

1.0%

.9%

.7%

.7%

.6%

.6%

.6%

.5%

.5%

.5%

.5%

.4%

22.4%

11.1%

10.8%

8.3%

7.1%

5.8%

2.3%

2.2%

2.2%

1.9%

1.8%

1.7%

1.5%

1.1%

1.1%

1.1%

1.1%

1.1%

1.4%

1.5%

1.0%

.0%

1.2%

1.0%

1.0%

.7%

1.1%

.9%

2.1%

1.9%

2.1%

2.5%

.7%

.7%

1.9%

22.8%

.0%

2.2%

11.4%

13.7%

5.7%

.3%

.0%

4.2%

1.7%

1.6%

1.0%

.7%

1.5%

2.5%

.0%

12.1%

33.6%

29.6%

2.7%

.0%

2.3%

3.8%

1.0%

.6%

5.5%

1.3%

.7%

.9%

.4%

.0%

.3%

.1%

.1%

.0%

.0%

1.3%

.2%

.1%

.0%

.0%

.3%

.2%

.0%

.0%

.0%

.4%

.0%

.0%

2.4%

1.6%

2.2%

1.3%

.0%

2.9%

.0%

.3%

.0%

.0%

1.6%

1.5%

.1%

.0%

.0%

.3%

1.5%

.0%

.0%

.0%

.4%

.0%

.0%

16.7%

25.2%

20.9%

4.7%

.0%

3.0%

3.9%

4.4%

.3%

2.8%

61

(Continued)

01991 Oataquest Incorporated June—Seproduction Prohibited

62

TABLE NUMBER:

T I T L E :

APPLICATION:

PLATFORM:

REGION:

UNITS:

Market Share Estimates—Electronic D e s ^ Automation Applications

22

1990 Final Market Share Estimates

Electronic CAE

Technical Workstation

Uorlduide

Millions of Dollars/Actual Units

Coopany

=======

AnaCAO

Aucotec

Quad Design Technology

EPIC Design Technology

Motorola

Quantic Laboratories

Tokyo Electron--NO OEM

<